Covariance functions for body weight from birth to maturity in Nellore cows.

PubMed

Boligon, A A; Mercadante, M E Z; Forni, S; Lôbo, R B; Albuquerque, L G

2010-03-01

The objective of this study was to estimate (co)variance functions using random regression models on Legendre polynomials for the analysis of repeated measures of BW from birth to adult age. A total of 82,064 records from 8,145 females were analyzed. Different models were compared. The models included additive direct and maternal effects, and animal and maternal permanent environmental effects as random terms. Contemporary group and dam age at calving (linear and quadratic effect) were included as fixed effects, and orthogonal Legendre polynomials of animal age (cubic regression) were considered as random covariables. Eight models with polynomials of third to sixth order were used to describe additive direct and maternal effects, and animal and maternal permanent environmental effects. Residual effects were modeled using 1 (i.e., assuming homogeneity of variances across all ages) or 5 age classes. The model with 5 classes was the best to describe the trajectory of residuals along the growth curve. The model including fourth- and sixth-order polynomials for additive direct and animal permanent environmental effects, respectively, and third-order polynomials for maternal genetic and maternal permanent environmental effects were the best. Estimates of (co)variance obtained with the multi-trait and random regression models were similar. Direct heritability estimates obtained with the random regression models followed a trend similar to that obtained with the multi-trait model. The largest estimates of maternal heritability were those of BW taken close to 240 d of age. In general, estimates of correlation between BW from birth to 8 yr of age decreased with increasing distance between ages.

The PX-EM algorithm for fast stable fitting of Henderson's mixed model

PubMed Central

Foulley, Jean-Louis; Van Dyk, David A

2000-01-01

This paper presents procedures for implementing the PX-EM algorithm of Liu, Rubin and Wu to compute REML estimates of variance covariance components in Henderson's linear mixed models. The class of models considered encompasses several correlated random factors having the same vector length e.g., as in random regression models for longitudinal data analysis and in sire-maternal grandsire models for genetic evaluation. Numerical examples are presented to illustrate the procedures. Much better results in terms of convergence characteristics (number of iterations and time required for convergence) are obtained for PX-EM relative to the basic EM algorithm in the random regression. PMID:14736399

Conditional Monte Carlo randomization tests for regression models.

PubMed

Parhat, Parwen; Rosenberger, William F; Diao, Guoqing

2014-08-15

We discuss the computation of randomization tests for clinical trials of two treatments when the primary outcome is based on a regression model. We begin by revisiting the seminal paper of Gail, Tan, and Piantadosi (1988), and then describe a method based on Monte Carlo generation of randomization sequences. The tests based on this Monte Carlo procedure are design based, in that they incorporate the particular randomization procedure used. We discuss permuted block designs, complete randomization, and biased coin designs. We also use a new technique by Plamadeala and Rosenberger (2012) for simple computation of conditional randomization tests. Like Gail, Tan, and Piantadosi, we focus on residuals from generalized linear models and martingale residuals from survival models. Such techniques do not apply to longitudinal data analysis, and we introduce a method for computation of randomization tests based on the predicted rate of change from a generalized linear mixed model when outcomes are longitudinal. We show, by simulation, that these randomization tests preserve the size and power well under model misspecification. Copyright © 2014 John Wiley & Sons, Ltd.

Random forest models to predict aqueous solubility.

PubMed

Palmer, David S; O'Boyle, Noel M; Glen, Robert C; Mitchell, John B O

2007-01-01

Random Forest regression (RF), Partial-Least-Squares (PLS) regression, Support Vector Machines (SVM), and Artificial Neural Networks (ANN) were used to develop QSPR models for the prediction of aqueous solubility, based on experimental data for 988 organic molecules. The Random Forest regression model predicted aqueous solubility more accurately than those created by PLS, SVM, and ANN and offered methods for automatic descriptor selection, an assessment of descriptor importance, and an in-parallel measure of predictive ability, all of which serve to recommend its use. The prediction of log molar solubility for an external test set of 330 molecules that are solid at 25 degrees C gave an r2 = 0.89 and RMSE = 0.69 log S units. For a standard data set selected from the literature, the model performed well with respect to other documented methods. Finally, the diversity of the training and test sets are compared to the chemical space occupied by molecules in the MDL drug data report, on the basis of molecular descriptors selected by the regression analysis.

Multiple-trait structured antedependence model to study the relationship between litter size and birth weight in pigs and rabbits.

PubMed

David, Ingrid; Garreau, Hervé; Balmisse, Elodie; Billon, Yvon; Canario, Laurianne

2017-01-20

Some genetic studies need to take into account correlations between traits that are repeatedly measured over time. Multiple-trait random regression models are commonly used to analyze repeated traits but suffer from several major drawbacks. In the present study, we developed a multiple-trait extension of the structured antedependence model (SAD) to overcome this issue and validated its usefulness by modeling the association between litter size (LS) and average birth weight (ABW) over parities in pigs and rabbits. The single-trait SAD model assumes that a random effect at time [Formula: see text] can be explained by the previous values of the random effect (i.e. at previous times). The proposed multiple-trait extension of the SAD model consists in adding a cross-antedependence parameter to the single-trait SAD model. This model can be easily fitted using ASReml and the OWN Fortran program that we have developed. In comparison with the random regression model, we used our multiple-trait SAD model to analyze the LS and ABW of 4345 litters from 1817 Large White sows and 8706 litters from 2286 L-1777 does over a maximum of five successive parities. For both species, the multiple-trait SAD fitted the data better than the random regression model. The difference between AIC of the two models (AIC_random regression-AIC_SAD) were equal to 7 and 227 for pigs and rabbits, respectively. A similar pattern of heritability and correlation estimates was obtained for both species. Heritabilities were lower for LS (ranging from 0.09 to 0.29) than for ABW (ranging from 0.23 to 0.39). The general trend was a decrease of the genetic correlation for a given trait between more distant parities. Estimates of genetic correlations between LS and ABW were negative and ranged from -0.03 to -0.52 across parities. No correlation was observed between the permanent environmental effects, except between the permanent environmental effects of LS and ABW of the same parity, for which the estimate of the correlation was strongly negative (ranging from -0.57 to -0.67). We demonstrated that application of our multiple-trait SAD model is feasible for studying several traits with repeated measurements and showed that it provided a better fit to the data than the random regression model.

Genetic improvement in mastitis resistance: comparison of selection criteria from cross-sectional and random regression sire models for somatic cell score.

PubMed

Odegård, J; Klemetsdal, G; Heringstad, B

2005-04-01

Several selection criteria for reducing incidence of mastitis were developed from a random regression sire model for test-day somatic cell score (SCS). For comparison, sire transmitting abilities were also predicted based on a cross-sectional model for lactation mean SCS. Only first-crop daughters were used in genetic evaluation of SCS, and the different selection criteria were compared based on their correlation with incidence of clinical mastitis in second-crop daughters (measured as mean daughter deviations). Selection criteria were predicted based on both complete and reduced first-crop daughter groups (261 or 65 daughters per sire, respectively). For complete daughter groups, predicted transmitting abilities at around 30 d in milk showed the best predictive ability for incidence of clinical mastitis, closely followed by average predicted transmitting abilities over the entire lactation. Both of these criteria were derived from the random regression model. These selection criteria improved accuracy of selection by approximately 2% relative to a cross-sectional model. However, for reduced daughter groups, the cross-sectional model yielded increased predictive ability compared with the selection criteria based on the random regression model. This result may be explained by the cross-sectional model being more robust, i.e., less sensitive to precision of (co)variance components estimates and effects of data structure.

Genetic analysis of body weights of individually fed beef bulls in South Africa using random regression models.

PubMed

Selapa, N W; Nephawe, K A; Maiwashe, A; Norris, D

2012-02-08

The aim of this study was to estimate genetic parameters for body weights of individually fed beef bulls measured at centralized testing stations in South Africa using random regression models. Weekly body weights of Bonsmara bulls (N = 2919) tested between 1999 and 2003 were available for the analyses. The model included a fixed regression of the body weights on fourth-order orthogonal Legendre polynomials of the actual days on test (7, 14, 21, 28, 35, 42, 49, 56, 63, 70, 77, and 84) for starting age and contemporary group effects. Random regressions on fourth-order orthogonal Legendre polynomials of the actual days on test were included for additive genetic effects and additional uncorrelated random effects of the weaning-herd-year and the permanent environment of the animal. Residual effects were assumed to be independently distributed with heterogeneous variance for each test day. Variance ratios for additive genetic, permanent environment and weaning-herd-year for weekly body weights at different test days ranged from 0.26 to 0.29, 0.37 to 0.44 and 0.26 to 0.34, respectively. The weaning-herd-year was found to have a significant effect on the variation of body weights of bulls despite a 28-day adjustment period. Genetic correlations amongst body weights at different test days were high, ranging from 0.89 to 1.00. Heritability estimates were comparable to literature using multivariate models. Therefore, random regression model could be applied in the genetic evaluation of body weight of individually fed beef bulls in South Africa.

Multiple Imputation of a Randomly Censored Covariate Improves Logistic Regression Analysis.

PubMed

Atem, Folefac D; Qian, Jing; Maye, Jacqueline E; Johnson, Keith A; Betensky, Rebecca A

2016-01-01

Randomly censored covariates arise frequently in epidemiologic studies. The most commonly used methods, including complete case and single imputation or substitution, suffer from inefficiency and bias. They make strong parametric assumptions or they consider limit of detection censoring only. We employ multiple imputation, in conjunction with semi-parametric modeling of the censored covariate, to overcome these shortcomings and to facilitate robust estimation. We develop a multiple imputation approach for randomly censored covariates within the framework of a logistic regression model. We use the non-parametric estimate of the covariate distribution or the semiparametric Cox model estimate in the presence of additional covariates in the model. We evaluate this procedure in simulations, and compare its operating characteristics to those from the complete case analysis and a survival regression approach. We apply the procedures to an Alzheimer's study of the association between amyloid positivity and maternal age of onset of dementia. Multiple imputation achieves lower standard errors and higher power than the complete case approach under heavy and moderate censoring and is comparable under light censoring. The survival regression approach achieves the highest power among all procedures, but does not produce interpretable estimates of association. Multiple imputation offers a favorable alternative to complete case analysis and ad hoc substitution methods in the presence of randomly censored covariates within the framework of logistic regression.

Patterns of medicinal plant use: an examination of the Ecuadorian Shuar medicinal flora using contingency table and binomial analyses.

PubMed

Bennett, Bradley C; Husby, Chad E

2008-03-28

Botanical pharmacopoeias are non-random subsets of floras, with some taxonomic groups over- or under-represented. Moerman [Moerman, D.E., 1979. Symbols and selectivity: a statistical analysis of Native American medical ethnobotany, Journal of Ethnopharmacology 1, 111-119] introduced linear regression/residual analysis to examine these patterns. However, regression, the commonly-employed analysis, suffers from several statistical flaws. We use contingency table and binomial analyses to examine patterns of Shuar medicinal plant use (from Amazonian Ecuador). We first analyzed the Shuar data using Moerman's approach, modified to better meet requirements of linear regression analysis. Second, we assessed the exact randomization contingency table test for goodness of fit. Third, we developed a binomial model to test for non-random selection of plants in individual families. Modified regression models (which accommodated assumptions of linear regression) reduced R(2) to from 0.59 to 0.38, but did not eliminate all problems associated with regression analyses. Contingency table analyses revealed that the entire flora departs from the null model of equal proportions of medicinal plants in all families. In the binomial analysis, only 10 angiosperm families (of 115) differed significantly from the null model. These 10 families are largely responsible for patterns seen at higher taxonomic levels. Contingency table and binomial analyses offer an easy and statistically valid alternative to the regression approach.

Parametric correlation functions to model the structure of permanent environmental (co)variances in milk yield random regression models.

PubMed

Bignardi, A B; El Faro, L; Cardoso, V L; Machado, P F; Albuquerque, L G

2009-09-01

The objective of the present study was to estimate milk yield genetic parameters applying random regression models and parametric correlation functions combined with a variance function to model animal permanent environmental effects. A total of 152,145 test-day milk yields from 7,317 first lactations of Holstein cows belonging to herds located in the southeastern region of Brazil were analyzed. Test-day milk yields were divided into 44 weekly classes of days in milk. Contemporary groups were defined by herd-test-day comprising a total of 2,539 classes. The model included direct additive genetic, permanent environmental, and residual random effects. The following fixed effects were considered: contemporary group, age of cow at calving (linear and quadratic regressions), and the population average lactation curve modeled by fourth-order orthogonal Legendre polynomial. Additive genetic effects were modeled by random regression on orthogonal Legendre polynomials of days in milk, whereas permanent environmental effects were estimated using a stationary or nonstationary parametric correlation function combined with a variance function of different orders. The structure of residual variances was modeled using a step function containing 6 variance classes. The genetic parameter estimates obtained with the model using a stationary correlation function associated with a variance function to model permanent environmental effects were similar to those obtained with models employing orthogonal Legendre polynomials for the same effect. A model using a sixth-order polynomial for additive effects and a stationary parametric correlation function associated with a seventh-order variance function to model permanent environmental effects would be sufficient for data fitting.

Random regression models using different functions to model milk flow in dairy cows.

PubMed

Laureano, M M M; Bignardi, A B; El Faro, L; Cardoso, V L; Tonhati, H; Albuquerque, L G

2014-09-12

We analyzed 75,555 test-day milk flow records from 2175 primiparous Holstein cows that calved between 1997 and 2005. Milk flow was obtained by dividing the mean milk yield (kg) of the 3 daily milking by the total milking time (min) and was expressed as kg/min. Milk flow was grouped into 43 weekly classes. The analyses were performed using a single-trait Random Regression Models that included direct additive genetic, permanent environmental, and residual random effects. In addition, the contemporary group and linear and quadratic effects of cow age at calving were included as fixed effects. Fourth-order orthogonal Legendre polynomial of days in milk was used to model the mean trend in milk flow. The additive genetic and permanent environmental covariance functions were estimated using random regression Legendre polynomials and B-spline functions of days in milk. The model using a third-order Legendre polynomial for additive genetic effects and a sixth-order polynomial for permanent environmental effects, which contained 7 residual classes, proved to be the most adequate to describe variations in milk flow, and was also the most parsimonious. The heritability in milk flow estimated by the most parsimonious model was of moderate to high magnitude.

Probability Theory Plus Noise: Descriptive Estimation and Inferential Judgment.

PubMed

Costello, Fintan; Watts, Paul

2018-01-01

We describe a computational model of two central aspects of people's probabilistic reasoning: descriptive probability estimation and inferential probability judgment. This model assumes that people's reasoning follows standard frequentist probability theory, but it is subject to random noise. This random noise has a regressive effect in descriptive probability estimation, moving probability estimates away from normative probabilities and toward the center of the probability scale. This random noise has an anti-regressive effect in inferential judgement, however. These regressive and anti-regressive effects explain various reliable and systematic biases seen in people's descriptive probability estimation and inferential probability judgment. This model predicts that these contrary effects will tend to cancel out in tasks that involve both descriptive estimation and inferential judgement, leading to unbiased responses in those tasks. We test this model by applying it to one such task, described by Gallistel et al. ). Participants' median responses in this task were unbiased, agreeing with normative probability theory over the full range of responses. Our model captures the pattern of unbiased responses in this task, while simultaneously explaining systematic biases away from normatively correct probabilities seen in other tasks. Copyright © 2018 Cognitive Science Society, Inc.

Learning accurate and interpretable models based on regularized random forests regression

PubMed Central

2014-01-01

Background Many biology related research works combine data from multiple sources in an effort to understand the underlying problems. It is important to find and interpret the most important information from these sources. Thus it will be beneficial to have an effective algorithm that can simultaneously extract decision rules and select critical features for good interpretation while preserving the prediction performance. Methods In this study, we focus on regression problems for biological data where target outcomes are continuous. In general, models constructed from linear regression approaches are relatively easy to interpret. However, many practical biological applications are nonlinear in essence where we can hardly find a direct linear relationship between input and output. Nonlinear regression techniques can reveal nonlinear relationship of data, but are generally hard for human to interpret. We propose a rule based regression algorithm that uses 1-norm regularized random forests. The proposed approach simultaneously extracts a small number of rules from generated random forests and eliminates unimportant features. Results We tested the approach on some biological data sets. The proposed approach is able to construct a significantly smaller set of regression rules using a subset of attributes while achieving prediction performance comparable to that of random forests regression. Conclusion It demonstrates high potential in aiding prediction and interpretation of nonlinear relationships of the subject being studied. PMID:25350120

Logistic Regression with Multiple Random Effects: A Simulation Study of Estimation Methods and Statistical Packages.

PubMed

Kim, Yoonsang; Choi, Young-Ku; Emery, Sherry

2013-08-01

Several statistical packages are capable of estimating generalized linear mixed models and these packages provide one or more of three estimation methods: penalized quasi-likelihood, Laplace, and Gauss-Hermite. Many studies have investigated these methods' performance for the mixed-effects logistic regression model. However, the authors focused on models with one or two random effects and assumed a simple covariance structure between them, which may not be realistic. When there are multiple correlated random effects in a model, the computation becomes intensive, and often an algorithm fails to converge. Moreover, in our analysis of smoking status and exposure to anti-tobacco advertisements, we have observed that when a model included multiple random effects, parameter estimates varied considerably from one statistical package to another even when using the same estimation method. This article presents a comprehensive review of the advantages and disadvantages of each estimation method. In addition, we compare the performances of the three methods across statistical packages via simulation, which involves two- and three-level logistic regression models with at least three correlated random effects. We apply our findings to a real dataset. Our results suggest that two packages-SAS GLIMMIX Laplace and SuperMix Gaussian quadrature-perform well in terms of accuracy, precision, convergence rates, and computing speed. We also discuss the strengths and weaknesses of the two packages in regard to sample sizes.

Logistic Regression with Multiple Random Effects: A Simulation Study of Estimation Methods and Statistical Packages

PubMed Central

Kim, Yoonsang; Emery, Sherry

2013-01-01

Several statistical packages are capable of estimating generalized linear mixed models and these packages provide one or more of three estimation methods: penalized quasi-likelihood, Laplace, and Gauss-Hermite. Many studies have investigated these methods’ performance for the mixed-effects logistic regression model. However, the authors focused on models with one or two random effects and assumed a simple covariance structure between them, which may not be realistic. When there are multiple correlated random effects in a model, the computation becomes intensive, and often an algorithm fails to converge. Moreover, in our analysis of smoking status and exposure to anti-tobacco advertisements, we have observed that when a model included multiple random effects, parameter estimates varied considerably from one statistical package to another even when using the same estimation method. This article presents a comprehensive review of the advantages and disadvantages of each estimation method. In addition, we compare the performances of the three methods across statistical packages via simulation, which involves two- and three-level logistic regression models with at least three correlated random effects. We apply our findings to a real dataset. Our results suggest that two packages—SAS GLIMMIX Laplace and SuperMix Gaussian quadrature—perform well in terms of accuracy, precision, convergence rates, and computing speed. We also discuss the strengths and weaknesses of the two packages in regard to sample sizes. PMID:24288415

Normalization Regression Estimation With Application to a Nonorthogonal, Nonrecursive Model of School Learning.

ERIC Educational Resources Information Center

Bulcock, J. W.; And Others

Advantages of normalization regression estimation over ridge regression estimation are demonstrated by reference to Bloom's model of school learning. Theoretical concern centered on the structure of scholastic achievement at grade 10 in Canadian high schools. Data on 886 students were randomly sampled from the Carnegie Human Resources Data Bank.…

Comparison of random regression test-day models for Polish Black and White cattle.

PubMed

Strabel, T; Szyda, J; Ptak, E; Jamrozik, J

2005-10-01

Test-day milk yields of first-lactation Black and White cows were used to select the model for routine genetic evaluation of dairy cattle in Poland. The population of Polish Black and White cows is characterized by small herd size, low level of production, and relatively early peak of lactation. Several random regression models for first-lactation milk yield were initially compared using the "percentage of squared bias" criterion and the correlations between true and predicted breeding values. Models with random herd-test-date effects, fixed age-season and herd-year curves, and random additive genetic and permanent environmental curves (Legendre polynomials of different orders were used for all regressions) were chosen for further studies. Additional comparisons included analyses of the residuals and shapes of variance curves in days in milk. The low production level and early peak of lactation of the breed required the use of Legendre polynomials of order 5 to describe age-season lactation curves. For the other curves, Legendre polynomials of order 3 satisfactorily described daily milk yield variation. Fitting third-order polynomials for the permanent environmental effect made it possible to adequately account for heterogeneous residual variance at different stages of lactation.

LiDAR based prediction of forest biomass using hierarchical models with spatially varying coefficients

USGS Publications Warehouse

Babcock, Chad; Finley, Andrew O.; Bradford, John B.; Kolka, Randall K.; Birdsey, Richard A.; Ryan, Michael G.

2015-01-01

Many studies and production inventory systems have shown the utility of coupling covariates derived from Light Detection and Ranging (LiDAR) data with forest variables measured on georeferenced inventory plots through regression models. The objective of this study was to propose and assess the use of a Bayesian hierarchical modeling framework that accommodates both residual spatial dependence and non-stationarity of model covariates through the introduction of spatial random effects. We explored this objective using four forest inventory datasets that are part of the North American Carbon Program, each comprising point-referenced measures of above-ground forest biomass and discrete LiDAR. For each dataset, we considered at least five regression model specifications of varying complexity. Models were assessed based on goodness of fit criteria and predictive performance using a 10-fold cross-validation procedure. Results showed that the addition of spatial random effects to the regression model intercept improved fit and predictive performance in the presence of substantial residual spatial dependence. Additionally, in some cases, allowing either some or all regression slope parameters to vary spatially, via the addition of spatial random effects, further improved model fit and predictive performance. In other instances, models showed improved fit but decreased predictive performance—indicating over-fitting and underscoring the need for cross-validation to assess predictive ability. The proposed Bayesian modeling framework provided access to pixel-level posterior predictive distributions that were useful for uncertainty mapping, diagnosing spatial extrapolation issues, revealing missing model covariates, and discovering locally significant parameters.

Genetic analysis of milk production traits of Tunisian Holsteins using random regression test-day model with Legendre polynomials

PubMed Central

2018-01-01

Objective The objective of this study was to estimate genetic parameters of milk, fat, and protein yields within and across lactations in Tunisian Holsteins using a random regression test-day (TD) model. Methods A random regression multiple trait multiple lactation TD model was used to estimate genetic parameters in the Tunisian dairy cattle population. Data were TD yields of milk, fat, and protein from the first three lactations. Random regressions were modeled with third-order Legendre polynomials for the additive genetic, and permanent environment effects. Heritabilities, and genetic correlations were estimated by Bayesian techniques using the Gibbs sampler. Results All variance components tended to be high in the beginning and the end of lactations. Additive genetic variances for milk, fat, and protein yields were the lowest and were the least variable compared to permanent variances. Heritability values tended to increase with parity. Estimates of heritabilities for 305-d yield-traits were low to moderate, 0.14 to 0.2, 0.12 to 0.17, and 0.13 to 0.18 for milk, fat, and protein yields, respectively. Within-parity, genetic correlations among traits were up to 0.74. Genetic correlations among lactations for the yield traits were relatively high and ranged from 0.78±0.01 to 0.82±0.03, between the first and second parities, from 0.73±0.03 to 0.8±0.04 between the first and third parities, and from 0.82±0.02 to 0.84±0.04 between the second and third parities. Conclusion These results are comparable to previously reported estimates on the same population, indicating that the adoption of a random regression TD model as the official genetic evaluation for production traits in Tunisia, as developed by most Interbull countries, is possible in the Tunisian Holsteins. PMID:28823122

Genetic analysis of milk production traits of Tunisian Holsteins using random regression test-day model with Legendre polynomials.

PubMed

Ben Zaabza, Hafedh; Ben Gara, Abderrahmen; Rekik, Boulbaba

2018-05-01

The objective of this study was to estimate genetic parameters of milk, fat, and protein yields within and across lactations in Tunisian Holsteins using a random regression test-day (TD) model. A random regression multiple trait multiple lactation TD model was used to estimate genetic parameters in the Tunisian dairy cattle population. Data were TD yields of milk, fat, and protein from the first three lactations. Random regressions were modeled with third-order Legendre polynomials for the additive genetic, and permanent environment effects. Heritabilities, and genetic correlations were estimated by Bayesian techniques using the Gibbs sampler. All variance components tended to be high in the beginning and the end of lactations. Additive genetic variances for milk, fat, and protein yields were the lowest and were the least variable compared to permanent variances. Heritability values tended to increase with parity. Estimates of heritabilities for 305-d yield-traits were low to moderate, 0.14 to 0.2, 0.12 to 0.17, and 0.13 to 0.18 for milk, fat, and protein yields, respectively. Within-parity, genetic correlations among traits were up to 0.74. Genetic correlations among lactations for the yield traits were relatively high and ranged from 0.78±0.01 to 0.82±0.03, between the first and second parities, from 0.73±0.03 to 0.8±0.04 between the first and third parities, and from 0.82±0.02 to 0.84±0.04 between the second and third parities. These results are comparable to previously reported estimates on the same population, indicating that the adoption of a random regression TD model as the official genetic evaluation for production traits in Tunisia, as developed by most Interbull countries, is possible in the Tunisian Holsteins.

Application of Machine-Learning Models to Predict Tacrolimus Stable Dose in Renal Transplant Recipients

NASA Astrophysics Data System (ADS)

Tang, Jie; Liu, Rong; Zhang, Yue-Li; Liu, Mou-Ze; Hu, Yong-Fang; Shao, Ming-Jie; Zhu, Li-Jun; Xin, Hua-Wen; Feng, Gui-Wen; Shang, Wen-Jun; Meng, Xiang-Guang; Zhang, Li-Rong; Ming, Ying-Zi; Zhang, Wei

2017-02-01

Tacrolimus has a narrow therapeutic window and considerable variability in clinical use. Our goal was to compare the performance of multiple linear regression (MLR) and eight machine learning techniques in pharmacogenetic algorithm-based prediction of tacrolimus stable dose (TSD) in a large Chinese cohort. A total of 1,045 renal transplant patients were recruited, 80% of which were randomly selected as the “derivation cohort” to develop dose-prediction algorithm, while the remaining 20% constituted the “validation cohort” to test the final selected algorithm. MLR, artificial neural network (ANN), regression tree (RT), multivariate adaptive regression splines (MARS), boosted regression tree (BRT), support vector regression (SVR), random forest regression (RFR), lasso regression (LAR) and Bayesian additive regression trees (BART) were applied and their performances were compared in this work. Among all the machine learning models, RT performed best in both derivation [0.71 (0.67-0.76)] and validation cohorts [0.73 (0.63-0.82)]. In addition, the ideal rate of RT was 4% higher than that of MLR. To our knowledge, this is the first study to use machine learning models to predict TSD, which will further facilitate personalized medicine in tacrolimus administration in the future.

Random regression analyses using B-spline functions to model growth of Nellore cattle.

PubMed

Boligon, A A; Mercadante, M E Z; Lôbo, R B; Baldi, F; Albuquerque, L G

2012-02-01

The objective of this study was to estimate (co)variance components using random regression on B-spline functions to weight records obtained from birth to adulthood. A total of 82 064 weight records of 8145 females obtained from the data bank of the Nellore Breeding Program (PMGRN/Nellore Brazil) which started in 1987, were used. The models included direct additive and maternal genetic effects and animal and maternal permanent environmental effects as random. Contemporary group and dam age at calving (linear and quadratic effect) were included as fixed effects, and orthogonal Legendre polynomials of age (cubic regression) were considered as random covariate. The random effects were modeled using B-spline functions considering linear, quadratic and cubic polynomials for each individual segment. Residual variances were grouped in five age classes. Direct additive genetic and animal permanent environmental effects were modeled using up to seven knots (six segments). A single segment with two knots at the end points of the curve was used for the estimation of maternal genetic and maternal permanent environmental effects. A total of 15 models were studied, with the number of parameters ranging from 17 to 81. The models that used B-splines were compared with multi-trait analyses with nine weight traits and to a random regression model that used orthogonal Legendre polynomials. A model fitting quadratic B-splines, with four knots or three segments for direct additive genetic effect and animal permanent environmental effect and two knots for maternal additive genetic effect and maternal permanent environmental effect, was the most appropriate and parsimonious model to describe the covariance structure of the data. Selection for higher weight, such as at young ages, should be performed taking into account an increase in mature cow weight. Particularly, this is important in most of Nellore beef cattle production systems, where the cow herd is maintained on range conditions. There is limited modification of the growth curve of Nellore cattle with respect to the aim of selecting them for rapid growth at young ages while maintaining constant adult weight.

Use of random regression to estimate genetic parameters of temperament across an age continuum in a crossbred cattle population.

PubMed

Littlejohn, B P; Riley, D G; Welsh, T H; Randel, R D; Willard, S T; Vann, R C

2018-05-12

The objective was to estimate genetic parameters of temperament in beef cattle across an age continuum. The population consisted predominantly of Brahman-British crossbred cattle. Temperament was quantified by: 1) pen score (PS), the reaction of a calf to a single experienced evaluator on a scale of 1 to 5 (1 = calm, 5 = excitable); 2) exit velocity (EV), the rate (m/sec) at which a calf traveled 1.83 m upon exiting a squeeze chute; and 3) temperament score (TS), the numerical average of PS and EV. Covariates included days of age and proportion of Bos indicus in the calf and dam. Random regression models included the fixed effects determined from the repeated measures models, except for calf age. Likelihood ratio tests were used to determine the most appropriate random structures. In repeated measures models, the proportion of Bos indicus in the calf was positively related with each calf temperament trait (0.41 ± 0.20, 0.85 ± 0.21, and 0.57 ± 0.18 for PS, EV, and TS, respectively; P < 0.01). There was an effect of contemporary group (combinations of season, year of birth, and management group) and dam age (P < 0.001) in all models. From repeated records analyses, estimates of heritability (h2) were 0.34 ± 0.04, 0.31 ± 0.04, and 0.39 ± 0.04, while estimates of permanent environmental variance as a proportion of the phenotypic variance (c2) were 0.30 ± 0.04, 0.31 ± 0.03, and 0.34 ± 0.04 for PS, EV, and TS, respectively. Quadratic additive genetic random regressions on Legendre polynomials of age were significant for all traits. Quadratic permanent environmental random regressions were significant for PS and TS, but linear permanent environmental random regressions were significant for EV. Random regression results suggested that these components change across the age dimension of these data. There appeared to be an increasing influence of permanent environmental effects and decreasing influence of additive genetic effects corresponding to increasing calf age for EV, and to a lesser extent for TS. Inherited temperament may be overcome by accumulating environmental stimuli with increases in age, especially after weaning.

Genetic analyses of protein yield in dairy cows applying random regression models with time-dependent and temperature x humidity-dependent covariates.

PubMed

Brügemann, K; Gernand, E; von Borstel, U U; König, S

2011-08-01

Data used in the present study included 1,095,980 first-lactation test-day records for protein yield of 154,880 Holstein cows housed on 196 large-scale dairy farms in Germany. Data were recorded between 2002 and 2009 and merged with meteorological data from public weather stations. The maximum distance between each farm and its corresponding weather station was 50 km. Hourly temperature-humidity indexes (THI) were calculated using the mean of hourly measurements of dry bulb temperature and relative humidity. On the phenotypic scale, an increase in THI was generally associated with a decrease in daily protein yield. For genetic analyses, a random regression model was applied using time-dependent (d in milk, DIM) and THI-dependent covariates. Additive genetic and permanent environmental effects were fitted with this random regression model and Legendre polynomials of order 3 for DIM and THI. In addition, the fixed curve was modeled with Legendre polynomials of order 3. Heterogeneous residuals were fitted by dividing DIM into 5 classes, and by dividing THI into 4 classes, resulting in 20 different classes. Additive genetic variances for daily protein yield decreased with increasing degrees of heat stress and were lowest at the beginning of lactation and at extreme THI. Due to higher additive genetic variances, slightly higher permanent environment variances, and similar residual variances, heritabilities were highest for low THI in combination with DIM at the end of lactation. Genetic correlations among individual values for THI were generally >0.90. These trends from the complex random regression model were verified by applying relatively simple bivariate animal models for protein yield measured in 2 THI environments; that is, defining a THI value of 60 as a threshold. These high correlations indicate the absence of any substantial genotype × environment interaction for protein yield. However, heritabilities and additive genetic variances from the random regression model tended to be slightly higher in the THI range corresponding to cows' comfort zone. Selecting such superior environments for progeny testing can contribute to an accurate genetic differentiation among selection candidates. Copyright © 2011 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

Visible and near infrared spectroscopy coupled to random forest to quantify some soil quality parameters

NASA Astrophysics Data System (ADS)

de Santana, Felipe Bachion; de Souza, André Marcelo; Poppi, Ronei Jesus

2018-02-01

This study evaluates the use of visible and near infrared spectroscopy (Vis-NIRS) combined with multivariate regression based on random forest to quantify some quality soil parameters. The parameters analyzed were soil cation exchange capacity (CEC), sum of exchange bases (SB), organic matter (OM), clay and sand present in the soils of several regions of Brazil. Current methods for evaluating these parameters are laborious, timely and require various wet analytical methods that are not adequate for use in precision agriculture, where faster and automatic responses are required. The random forest regression models were statistically better than PLS regression models for CEC, OM, clay and sand, demonstrating resistance to overfitting, attenuating the effect of outlier samples and indicating the most important variables for the model. The methodology demonstrates the potential of the Vis-NIR as an alternative for determination of CEC, SB, OM, sand and clay, making possible to develop a fast and automatic analytical procedure.

An optimal sample data usage strategy to minimize overfitting and underfitting effects in regression tree models based on remotely-sensed data

USGS Publications Warehouse

Gu, Yingxin; Wylie, Bruce K.; Boyte, Stephen; Picotte, Joshua J.; Howard, Danny; Smith, Kelcy; Nelson, Kurtis

2016-01-01

Regression tree models have been widely used for remote sensing-based ecosystem mapping. Improper use of the sample data (model training and testing data) may cause overfitting and underfitting effects in the model. The goal of this study is to develop an optimal sampling data usage strategy for any dataset and identify an appropriate number of rules in the regression tree model that will improve its accuracy and robustness. Landsat 8 data and Moderate-Resolution Imaging Spectroradiometer-scaled Normalized Difference Vegetation Index (NDVI) were used to develop regression tree models. A Python procedure was designed to generate random replications of model parameter options across a range of model development data sizes and rule number constraints. The mean absolute difference (MAD) between the predicted and actual NDVI (scaled NDVI, value from 0–200) and its variability across the different randomized replications were calculated to assess the accuracy and stability of the models. In our case study, a six-rule regression tree model developed from 80% of the sample data had the lowest MAD (MADtraining = 2.5 and MADtesting = 2.4), which was suggested as the optimal model. This study demonstrates how the training data and rule number selections impact model accuracy and provides important guidance for future remote-sensing-based ecosystem modeling.

A general framework for the use of logistic regression models in meta-analysis.

PubMed

Simmonds, Mark C; Higgins, Julian Pt

2016-12-01

Where individual participant data are available for every randomised trial in a meta-analysis of dichotomous event outcomes, "one-stage" random-effects logistic regression models have been proposed as a way to analyse these data. Such models can also be used even when individual participant data are not available and we have only summary contingency table data. One benefit of this one-stage regression model over conventional meta-analysis methods is that it maximises the correct binomial likelihood for the data and so does not require the common assumption that effect estimates are normally distributed. A second benefit of using this model is that it may be applied, with only minor modification, in a range of meta-analytic scenarios, including meta-regression, network meta-analyses and meta-analyses of diagnostic test accuracy. This single model can potentially replace the variety of often complex methods used in these areas. This paper considers, with a range of meta-analysis examples, how random-effects logistic regression models may be used in a number of different types of meta-analyses. This one-stage approach is compared with widely used meta-analysis methods including Bayesian network meta-analysis and the bivariate and hierarchical summary receiver operating characteristic (ROC) models for meta-analyses of diagnostic test accuracy. © The Author(s) 2014.

3D statistical shape models incorporating 3D random forest regression voting for robust CT liver segmentation

NASA Astrophysics Data System (ADS)

Norajitra, Tobias; Meinzer, Hans-Peter; Maier-Hein, Klaus H.

2015-03-01

During image segmentation, 3D Statistical Shape Models (SSM) usually conduct a limited search for target landmarks within one-dimensional search profiles perpendicular to the model surface. In addition, landmark appearance is modeled only locally based on linear profiles and weak learners, altogether leading to segmentation errors from landmark ambiguities and limited search coverage. We present a new method for 3D SSM segmentation based on 3D Random Forest Regression Voting. For each surface landmark, a Random Regression Forest is trained that learns a 3D spatial displacement function between the according reference landmark and a set of surrounding sample points, based on an infinite set of non-local randomized 3D Haar-like features. Landmark search is then conducted omni-directionally within 3D search spaces, where voxelwise forest predictions on landmark position contribute to a common voting map which reflects the overall position estimate. Segmentation experiments were conducted on a set of 45 CT volumes of the human liver, of which 40 images were randomly chosen for training and 5 for testing. Without parameter optimization, using a simple candidate selection and a single resolution approach, excellent results were achieved, while faster convergence and better concavity segmentation were observed, altogether underlining the potential of our approach in terms of increased robustness from distinct landmark detection and from better search coverage.

Simple, Efficient Estimators of Treatment Effects in Randomized Trials Using Generalized Linear Models to Leverage Baseline Variables

PubMed Central

Rosenblum, Michael; van der Laan, Mark J.

2010-01-01

Models, such as logistic regression and Poisson regression models, are often used to estimate treatment effects in randomized trials. These models leverage information in variables collected before randomization, in order to obtain more precise estimates of treatment effects. However, there is the danger that model misspecification will lead to bias. We show that certain easy to compute, model-based estimators are asymptotically unbiased even when the working model used is arbitrarily misspecified. Furthermore, these estimators are locally efficient. As a special case of our main result, we consider a simple Poisson working model containing only main terms; in this case, we prove the maximum likelihood estimate of the coefficient corresponding to the treatment variable is an asymptotically unbiased estimator of the marginal log rate ratio, even when the working model is arbitrarily misspecified. This is the log-linear analog of ANCOVA for linear models. Our results demonstrate one application of targeted maximum likelihood estimation. PMID:20628636

Empirical likelihood inference in randomized clinical trials.

PubMed

Zhang, Biao

2017-01-01

In individually randomized controlled trials, in addition to the primary outcome, information is often available on a number of covariates prior to randomization. This information is frequently utilized to undertake adjustment for baseline characteristics in order to increase precision of the estimation of average treatment effects; such adjustment is usually performed via covariate adjustment in outcome regression models. Although the use of covariate adjustment is widely seen as desirable for making treatment effect estimates more precise and the corresponding hypothesis tests more powerful, there are considerable concerns that objective inference in randomized clinical trials can potentially be compromised. In this paper, we study an empirical likelihood approach to covariate adjustment and propose two unbiased estimating functions that automatically decouple evaluation of average treatment effects from regression modeling of covariate-outcome relationships. The resulting empirical likelihood estimator of the average treatment effect is as efficient as the existing efficient adjusted estimators 1 when separate treatment-specific working regression models are correctly specified, yet are at least as efficient as the existing efficient adjusted estimators 1 for any given treatment-specific working regression models whether or not they coincide with the true treatment-specific covariate-outcome relationships. We present a simulation study to compare the finite sample performance of various methods along with some results on analysis of a data set from an HIV clinical trial. The simulation results indicate that the proposed empirical likelihood approach is more efficient and powerful than its competitors when the working covariate-outcome relationships by treatment status are misspecified.

Novel solutions for an old disease: diagnosis of acute appendicitis with random forest, support vector machines, and artificial neural networks.

PubMed

Hsieh, Chung-Ho; Lu, Ruey-Hwa; Lee, Nai-Hsin; Chiu, Wen-Ta; Hsu, Min-Huei; Li, Yu-Chuan Jack

2011-01-01

Diagnosing acute appendicitis clinically is still difficult. We developed random forests, support vector machines, and artificial neural network models to diagnose acute appendicitis. Between January 2006 and December 2008, patients who had a consultation session with surgeons for suspected acute appendicitis were enrolled. Seventy-five percent of the data set was used to construct models including random forest, support vector machines, artificial neural networks, and logistic regression. Twenty-five percent of the data set was withheld to evaluate model performance. The area under the receiver operating characteristic curve (AUC) was used to evaluate performance, which was compared with that of the Alvarado score. Data from a total of 180 patients were collected, 135 used for training and 45 for testing. The mean age of patients was 39.4 years (range, 16-85). Final diagnosis revealed 115 patients with and 65 without appendicitis. The AUC of random forest, support vector machines, artificial neural networks, logistic regression, and Alvarado was 0.98, 0.96, 0.91, 0.87, and 0.77, respectively. The sensitivity, specificity, positive, and negative predictive values of random forest were 94%, 100%, 100%, and 87%, respectively. Random forest performed better than artificial neural networks, logistic regression, and Alvarado. We demonstrated that random forest can predict acute appendicitis with good accuracy and, deployed appropriately, can be an effective tool in clinical decision making. Copyright © 2011 Mosby, Inc. All rights reserved.

Comparative analysis of used car price evaluation models

NASA Astrophysics Data System (ADS)

Chen, Chuancan; Hao, Lulu; Xu, Cong

2017-05-01

An accurate used car price evaluation is a catalyst for the healthy development of used car market. Data mining has been applied to predict used car price in several articles. However, little is studied on the comparison of using different algorithms in used car price estimation. This paper collects more than 100,000 used car dealing records throughout China to do empirical analysis on a thorough comparison of two algorithms: linear regression and random forest. These two algorithms are used to predict used car price in three different models: model for a certain car make, model for a certain car series and universal model. Results show that random forest has a stable but not ideal effect in price evaluation model for a certain car make, but it shows great advantage in the universal model compared with linear regression. This indicates that random forest is an optimal algorithm when handling complex models with a large number of variables and samples, yet it shows no obvious advantage when coping with simple models with less variables.

Estimating the impact of mineral aerosols on crop yields in food insecure regions using statistical crop models

NASA Astrophysics Data System (ADS)

Hoffman, A.; Forest, C. E.; Kemanian, A.

2016-12-01

A significant number of food-insecure nations exist in regions of the world where dust plays a large role in the climate system. While the impacts of common climate variables (e.g. temperature, precipitation, ozone, and carbon dioxide) on crop yields are relatively well understood, the impact of mineral aerosols on yields have not yet been thoroughly investigated. This research aims to develop the data and tools to progress our understanding of mineral aerosol impacts on crop yields. Suspended dust affects crop yields by altering the amount and type of radiation reaching the plant, modifying local temperature and precipitation. While dust events (i.e. dust storms) affect crop yields by depleting the soil of nutrients or by defoliation via particle abrasion. The impact of dust on yields is modeled statistically because we are uncertain which impacts will dominate the response on national and regional scales considered in this study. Multiple linear regression is used in a number of large-scale statistical crop modeling studies to estimate yield responses to various climate variables. In alignment with previous work, we develop linear crop models, but build upon this simple method of regression with machine-learning techniques (e.g. random forests) to identify important statistical predictors and isolate how dust affects yields on the scales of interest. To perform this analysis, we develop a crop-climate dataset for maize, soybean, groundnut, sorghum, rice, and wheat for the regions of West Africa, East Africa, South Africa, and the Sahel. Random forest regression models consistently model historic crop yields better than the linear models. In several instances, the random forest models accurately capture the temperature and precipitation threshold behavior in crops. Additionally, improving agricultural technology has caused a well-documented positive trend that dominates time series of global and regional yields. This trend is often removed before regression with traditional crop models, but likely at the cost of removing climate information. Our random forest models consistently discover the positive trend without removing any additional data. The application of random forests as a statistical crop model provides insight into understanding the impact of dust on yields in marginal food producing regions.

Complementary nonparametric analysis of covariance for logistic regression in a randomized clinical trial setting.

PubMed

Tangen, C M; Koch, G G

1999-03-01

In the randomized clinical trial setting, controlling for covariates is expected to produce variance reduction for the treatment parameter estimate and to adjust for random imbalances of covariates between the treatment groups. However, for the logistic regression model, variance reduction is not obviously obtained. This can lead to concerns about the assumptions of the logistic model. We introduce a complementary nonparametric method for covariate adjustment. It provides results that are usually compatible with expectations for analysis of covariance. The only assumptions required are based on randomization and sampling arguments. The resulting treatment parameter is a (unconditional) population average log-odds ratio that has been adjusted for random imbalance of covariates. Data from a randomized clinical trial are used to compare results from the traditional maximum likelihood logistic method with those from the nonparametric logistic method. We examine treatment parameter estimates, corresponding standard errors, and significance levels in models with and without covariate adjustment. In addition, we discuss differences between unconditional population average treatment parameters and conditional subpopulation average treatment parameters. Additional features of the nonparametric method, including stratified (multicenter) and multivariate (multivisit) analyses, are illustrated. Extensions of this methodology to the proportional odds model are also made.

New machine learning tools for predictive vegetation mapping after climate change: Bagging and Random Forest perform better than Regression Tree Analysis

Treesearch

L.R. Iverson; A.M. Prasad; A. Liaw

2004-01-01

More and better machine learning tools are becoming available for landscape ecologists to aid in understanding species-environment relationships and to map probable species occurrence now and potentially into the future. To thal end, we evaluated three statistical models: Regression Tree Analybib (RTA), Bagging Trees (BT) and Random Forest (RF) for their utility in...

An application of quantile random forests for predictive mapping of forest attributes

Treesearch

E.A. Freeman; G.G. Moisen

2015-01-01

Increasingly, random forest models are used in predictive mapping of forest attributes. Traditional random forests output the mean prediction from the random trees. Quantile regression forests (QRF) is an extension of random forests developed by Nicolai Meinshausen that provides non-parametric estimates of the median predicted value as well as prediction quantiles. It...

Multiple-trait random regression models for the estimation of genetic parameters for milk, fat, and protein yield in buffaloes.

PubMed

Borquis, Rusbel Raul Aspilcueta; Neto, Francisco Ribeiro de Araujo; Baldi, Fernando; Hurtado-Lugo, Naudin; de Camargo, Gregório M F; Muñoz-Berrocal, Milthon; Tonhati, Humberto

2013-09-01

In this study, genetic parameters for test-day milk, fat, and protein yield were estimated for the first lactation. The data analyzed consisted of 1,433 first lactations of Murrah buffaloes, daughters of 113 sires from 12 herds in the state of São Paulo, Brazil, with calvings from 1985 to 2007. Ten-month classes of lactation days were considered for the test-day yields. The (co)variance components for the 3 traits were estimated using the regression analyses by Bayesian inference applying an animal model by Gibbs sampling. The contemporary groups were defined as herd-year-month of the test day. In the model, the random effects were additive genetic, permanent environment, and residual. The fixed effects were contemporary group and number of milkings (1 or 2), the linear and quadratic effects of the covariable age of the buffalo at calving, as well as the mean lactation curve of the population, which was modeled by orthogonal Legendre polynomials of fourth order. The random effects for the traits studied were modeled by Legendre polynomials of third and fourth order for additive genetic and permanent environment, respectively, the residual variances were modeled considering 4 residual classes. The heritability estimates for the traits were moderate (from 0.21-0.38), with higher estimates in the intermediate lactation phase. The genetic correlation estimates within and among the traits varied from 0.05 to 0.99. The results indicate that the selection for any trait test day will result in an indirect genetic gain for milk, fat, and protein yield in all periods of the lactation curve. The accuracy associated with estimated breeding values obtained using multi-trait random regression was slightly higher (around 8%) compared with single-trait random regression. This difference may be because to the greater amount of information available per animal. Copyright © 2013 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

Longitudinal analysis of the strengths and difficulties questionnaire scores of the Millennium Cohort Study children in England using M-quantile random-effects regression.

PubMed

Tzavidis, Nikos; Salvati, Nicola; Schmid, Timo; Flouri, Eirini; Midouhas, Emily

2016-02-01

Multilevel modelling is a popular approach for longitudinal data analysis. Statistical models conventionally target a parameter at the centre of a distribution. However, when the distribution of the data is asymmetric, modelling other location parameters, e.g. percentiles, may be more informative. We present a new approach, M -quantile random-effects regression, for modelling multilevel data. The proposed method is used for modelling location parameters of the distribution of the strengths and difficulties questionnaire scores of children in England who participate in the Millennium Cohort Study. Quantile mixed models are also considered. The analyses offer insights to child psychologists about the differential effects of risk factors on children's outcomes.

Forecasting Daily Patient Outflow From a Ward Having No Real-Time Clinical Data

PubMed Central

Tran, Truyen; Luo, Wei; Phung, Dinh; Venkatesh, Svetha

2016-01-01

Background: Modeling patient flow is crucial in understanding resource demand and prioritization. We study patient outflow from an open ward in an Australian hospital, where currently bed allocation is carried out by a manager relying on past experiences and looking at demand. Automatic methods that provide a reasonable estimate of total next-day discharges can aid in efficient bed management. The challenges in building such methods lie in dealing with large amounts of discharge noise introduced by the nonlinear nature of hospital procedures, and the nonavailability of real-time clinical information in wards. Objective Our study investigates different models to forecast the total number of next-day discharges from an open ward having no real-time clinical data. Methods We compared 5 popular regression algorithms to model total next-day discharges: (1) autoregressive integrated moving average (ARIMA), (2) the autoregressive moving average with exogenous variables (ARMAX), (3) k-nearest neighbor regression, (4) random forest regression, and (5) support vector regression. Although the autoregressive integrated moving average model relied on past 3-month discharges, nearest neighbor forecasting used median of similar discharges in the past in estimating next-day discharge. In addition, the ARMAX model used the day of the week and number of patients currently in ward as exogenous variables. For the random forest and support vector regression models, we designed a predictor set of 20 patient features and 88 ward-level features. Results Our data consisted of 12,141 patient visits over 1826 days. Forecasting quality was measured using mean forecast error, mean absolute error, symmetric mean absolute percentage error, and root mean square error. When compared with a moving average prediction model, all 5 models demonstrated superior performance with the random forests achieving 22.7% improvement in mean absolute error, for all days in the year 2014. Conclusions In the absence of clinical information, our study recommends using patient-level and ward-level data in predicting next-day discharges. Random forest and support vector regression models are able to use all available features from such data, resulting in superior performance over traditional autoregressive methods. An intelligent estimate of available beds in wards plays a crucial role in relieving access block in emergency departments. PMID:27444059

Logistic regression of family data from retrospective study designs.

PubMed

Whittemore, Alice S; Halpern, Jerry

2003-11-01

We wish to study the effects of genetic and environmental factors on disease risk, using data from families ascertained because they contain multiple cases of the disease. To do so, we must account for the way participants were ascertained, and for within-family correlations in both disease occurrences and covariates. We model the joint probability distribution of the covariates of ascertained family members, given family disease occurrence and pedigree structure. We describe two such covariate models: the random effects model and the marginal model. Both models assume a logistic form for the distribution of one person's covariates that involves a vector beta of regression parameters. The components of beta in the two models have different interpretations, and they differ in magnitude when the covariates are correlated within families. We describe ascertainment assumptions needed to estimate consistently the parameters beta(RE) in the random effects model and the parameters beta(M) in the marginal model. Under the ascertainment assumptions for the random effects model, we show that conditional logistic regression (CLR) of matched family data gives a consistent estimate beta(RE) for beta(RE) and a consistent estimate for the covariance matrix of beta(RE). Under the ascertainment assumptions for the marginal model, we show that unconditional logistic regression (ULR) gives a consistent estimate for beta(M), and we give a consistent estimator for its covariance matrix. The random effects/CLR approach is simple to use and to interpret, but it can use data only from families containing both affected and unaffected members. The marginal/ULR approach uses data from all individuals, but its variance estimates require special computations. A C program to compute these variance estimates is available at http://www.stanford.edu/dept/HRP/epidemiology. We illustrate these pros and cons by application to data on the effects of parity on ovarian cancer risk in mother/daughter pairs, and use simulations to study the performance of the estimates. Copyright 2003 Wiley-Liss, Inc.

Comparisons between physics-based, engineering, and statistical learning models for outdoor sound propagation.

PubMed

Hart, Carl R; Reznicek, Nathan J; Wilson, D Keith; Pettit, Chris L; Nykaza, Edward T

2016-05-01

Many outdoor sound propagation models exist, ranging from highly complex physics-based simulations to simplified engineering calculations, and more recently, highly flexible statistical learning methods. Several engineering and statistical learning models are evaluated by using a particular physics-based model, namely, a Crank-Nicholson parabolic equation (CNPE), as a benchmark. Narrowband transmission loss values predicted with the CNPE, based upon a simulated data set of meteorological, boundary, and source conditions, act as simulated observations. In the simulated data set sound propagation conditions span from downward refracting to upward refracting, for acoustically hard and soft boundaries, and low frequencies. Engineering models used in the comparisons include the ISO 9613-2 method, Harmonoise, and Nord2000 propagation models. Statistical learning methods used in the comparisons include bagged decision tree regression, random forest regression, boosting regression, and artificial neural network models. Computed skill scores are relative to sound propagation in a homogeneous atmosphere over a rigid ground. Overall skill scores for the engineering noise models are 0.6%, -7.1%, and 83.8% for the ISO 9613-2, Harmonoise, and Nord2000 models, respectively. Overall skill scores for the statistical learning models are 99.5%, 99.5%, 99.6%, and 99.6% for bagged decision tree, random forest, boosting, and artificial neural network regression models, respectively.

Revisiting Fixed- and Random-Effects Models: Some Considerations for Policy-Relevant Education Research

ERIC Educational Resources Information Center

Clarke, Paul; Crawford, Claire; Steele, Fiona; Vignoles, Anna

2015-01-01

The use of fixed (FE) and random effects (RE) in two-level hierarchical linear regression is discussed in the context of education research. We compare the robustness of FE models with the modelling flexibility and potential efficiency of those from RE models. We argue that the two should be seen as complementary approaches. We then compare both…

Random regression models on Legendre polynomials to estimate genetic parameters for weights from birth to adult age in Canchim cattle.

PubMed

Baldi, F; Albuquerque, L G; Alencar, M M

2010-08-01

The objective of this work was to estimate covariance functions for direct and maternal genetic effects, animal and maternal permanent environmental effects, and subsequently, to derive relevant genetic parameters for growth traits in Canchim cattle. Data comprised 49,011 weight records on 2435 females from birth to adult age. The model of analysis included fixed effects of contemporary groups (year and month of birth and at weighing) and age of dam as quadratic covariable. Mean trends were taken into account by a cubic regression on orthogonal polynomials of animal age. Residual variances were allowed to vary and were modelled by a step function with 1, 4 or 11 classes based on animal's age. The model fitting four classes of residual variances was the best. A total of 12 random regression models from second to seventh order were used to model direct and maternal genetic effects, animal and maternal permanent environmental effects. The model with direct and maternal genetic effects, animal and maternal permanent environmental effects fitted by quadric, cubic, quintic and linear Legendre polynomials, respectively, was the most adequate to describe the covariance structure of the data. Estimates of direct and maternal heritability obtained by multi-trait (seven traits) and random regression models were very similar. Selection for higher weight at any age, especially after weaning, will produce an increase in mature cow weight. The possibility to modify the growth curve in Canchim cattle to obtain animals with rapid growth at early ages and moderate to low mature cow weight is limited.

Robust, Adaptive Functional Regression in Functional Mixed Model Framework.

PubMed

Zhu, Hongxiao; Brown, Philip J; Morris, Jeffrey S

2011-09-01

Functional data are increasingly encountered in scientific studies, and their high dimensionality and complexity lead to many analytical challenges. Various methods for functional data analysis have been developed, including functional response regression methods that involve regression of a functional response on univariate/multivariate predictors with nonparametrically represented functional coefficients. In existing methods, however, the functional regression can be sensitive to outlying curves and outlying regions of curves, so is not robust. In this paper, we introduce a new Bayesian method, robust functional mixed models (R-FMM), for performing robust functional regression within the general functional mixed model framework, which includes multiple continuous or categorical predictors and random effect functions accommodating potential between-function correlation induced by the experimental design. The underlying model involves a hierarchical scale mixture model for the fixed effects, random effect and residual error functions. These modeling assumptions across curves result in robust nonparametric estimators of the fixed and random effect functions which down-weight outlying curves and regions of curves, and produce statistics that can be used to flag global and local outliers. These assumptions also lead to distributions across wavelet coefficients that have outstanding sparsity and adaptive shrinkage properties, with great flexibility for the data to determine the sparsity and the heaviness of the tails. Together with the down-weighting of outliers, these within-curve properties lead to fixed and random effect function estimates that appear in our simulations to be remarkably adaptive in their ability to remove spurious features yet retain true features of the functions. We have developed general code to implement this fully Bayesian method that is automatic, requiring the user to only provide the functional data and design matrices. It is efficient enough to handle large data sets, and yields posterior samples of all model parameters that can be used to perform desired Bayesian estimation and inference. Although we present details for a specific implementation of the R-FMM using specific distributional choices in the hierarchical model, 1D functions, and wavelet transforms, the method can be applied more generally using other heavy-tailed distributions, higher dimensional functions (e.g. images), and using other invertible transformations as alternatives to wavelets.

Robust, Adaptive Functional Regression in Functional Mixed Model Framework

PubMed Central

Zhu, Hongxiao; Brown, Philip J.; Morris, Jeffrey S.

2012-01-01

Functional data are increasingly encountered in scientific studies, and their high dimensionality and complexity lead to many analytical challenges. Various methods for functional data analysis have been developed, including functional response regression methods that involve regression of a functional response on univariate/multivariate predictors with nonparametrically represented functional coefficients. In existing methods, however, the functional regression can be sensitive to outlying curves and outlying regions of curves, so is not robust. In this paper, we introduce a new Bayesian method, robust functional mixed models (R-FMM), for performing robust functional regression within the general functional mixed model framework, which includes multiple continuous or categorical predictors and random effect functions accommodating potential between-function correlation induced by the experimental design. The underlying model involves a hierarchical scale mixture model for the fixed effects, random effect and residual error functions. These modeling assumptions across curves result in robust nonparametric estimators of the fixed and random effect functions which down-weight outlying curves and regions of curves, and produce statistics that can be used to flag global and local outliers. These assumptions also lead to distributions across wavelet coefficients that have outstanding sparsity and adaptive shrinkage properties, with great flexibility for the data to determine the sparsity and the heaviness of the tails. Together with the down-weighting of outliers, these within-curve properties lead to fixed and random effect function estimates that appear in our simulations to be remarkably adaptive in their ability to remove spurious features yet retain true features of the functions. We have developed general code to implement this fully Bayesian method that is automatic, requiring the user to only provide the functional data and design matrices. It is efficient enough to handle large data sets, and yields posterior samples of all model parameters that can be used to perform desired Bayesian estimation and inference. Although we present details for a specific implementation of the R-FMM using specific distributional choices in the hierarchical model, 1D functions, and wavelet transforms, the method can be applied more generally using other heavy-tailed distributions, higher dimensional functions (e.g. images), and using other invertible transformations as alternatives to wavelets. PMID:22308015

Modeling urban coastal flood severity from crowd-sourced flood reports using Poisson regression and Random Forest

NASA Astrophysics Data System (ADS)

Sadler, J. M.; Goodall, J. L.; Morsy, M. M.; Spencer, K.

2018-04-01

Sea level rise has already caused more frequent and severe coastal flooding and this trend will likely continue. Flood prediction is an essential part of a coastal city's capacity to adapt to and mitigate this growing problem. Complex coastal urban hydrological systems however, do not always lend themselves easily to physically-based flood prediction approaches. This paper presents a method for using a data-driven approach to estimate flood severity in an urban coastal setting using crowd-sourced data, a non-traditional but growing data source, along with environmental observation data. Two data-driven models, Poisson regression and Random Forest regression, are trained to predict the number of flood reports per storm event as a proxy for flood severity, given extensive environmental data (i.e., rainfall, tide, groundwater table level, and wind conditions) as input. The method is demonstrated using data from Norfolk, Virginia USA from September 2010 to October 2016. Quality-controlled, crowd-sourced street flooding reports ranging from 1 to 159 per storm event for 45 storm events are used to train and evaluate the models. Random Forest performed better than Poisson regression at predicting the number of flood reports and had a lower false negative rate. From the Random Forest model, total cumulative rainfall was by far the most dominant input variable in predicting flood severity, followed by low tide and lower low tide. These methods serve as a first step toward using data-driven methods for spatially and temporally detailed coastal urban flood prediction.

Prediction of random-regression coefficient for daily milk yield after 305 days in milk by using the regression-coefficient estimates from the first 305 days.

PubMed

Yamazaki, Takeshi; Takeda, Hisato; Hagiya, Koichi; Yamaguchi, Satoshi; Sasaki, Osamu

2018-03-13

Because lactation periods in dairy cows lengthen with increasing total milk production, it is important to predict individual productivities after 305 days in milk (DIM) to determine the optimal lactation period. We therefore examined whether the random regression (RR) coefficient from 306 to 450 DIM (M2) can be predicted from those during the first 305 DIM (M1) by using a random regression model. We analyzed test-day milk records from 85690 Holstein cows in their first lactations and 131727 cows in their later (second to fifth) lactations. Data in M1 and M2 were analyzed separately by using different single-trait RR animal models. We then performed a multiple regression analysis of the RR coefficients of M2 on those of M1 during the first and later lactations. The first-order Legendre polynomials were practical covariates of random regression for the milk yields of M2. All RR coefficients for the additive genetic (AG) effect and the intercept for the permanent environmental (PE) effect of M2 had moderate to strong correlations with the intercept for the AG effect of M1. The coefficients of determination for multiple regression of the combined intercepts for the AG and PE effects of M2 on the coefficients for the AG effect of M1 were moderate to high. The daily milk yields of M2 predicted by using the RR coefficients for the AG effect of M1 were highly correlated with those obtained by using the coefficients of M2. Milk production after 305 DIM can be predicted by using the RR coefficient estimates of the AG effect during the first 305 DIM.

Bayesian hierarchical models for cost-effectiveness analyses that use data from cluster randomized trials.

PubMed

Grieve, Richard; Nixon, Richard; Thompson, Simon G

2010-01-01

Cost-effectiveness analyses (CEA) may be undertaken alongside cluster randomized trials (CRTs) where randomization is at the level of the cluster (for example, the hospital or primary care provider) rather than the individual. Costs (and outcomes) within clusters may be correlated so that the assumption made by standard bivariate regression models, that observations are independent, is incorrect. This study develops a flexible modeling framework to acknowledge the clustering in CEA that use CRTs. The authors extend previous Bayesian bivariate models for CEA of multicenter trials to recognize the specific form of clustering in CRTs. They develop new Bayesian hierarchical models (BHMs) that allow mean costs and outcomes, and also variances, to differ across clusters. They illustrate how each model can be applied using data from a large (1732 cases, 70 primary care providers) CRT evaluating alternative interventions for reducing postnatal depression. The analyses compare cost-effectiveness estimates from BHMs with standard bivariate regression models that ignore the data hierarchy. The BHMs show high levels of cost heterogeneity across clusters (intracluster correlation coefficient, 0.17). Compared with standard regression models, the BHMs yield substantially increased uncertainty surrounding the cost-effectiveness estimates, and altered point estimates. The authors conclude that ignoring clustering can lead to incorrect inferences. The BHMs that they present offer a flexible modeling framework that can be applied more generally to CEA that use CRTs.

Regression discontinuity was a valid design for dichotomous outcomes in three randomized trials.

PubMed

van Leeuwen, Nikki; Lingsma, Hester F; Mooijaart, Simon P; Nieboer, Daan; Trompet, Stella; Steyerberg, Ewout W

2018-06-01

Regression discontinuity (RD) is a quasi-experimental design that may provide valid estimates of treatment effects in case of continuous outcomes. We aimed to evaluate validity and precision in the RD design for dichotomous outcomes. We performed validation studies in three large randomized controlled trials (RCTs) (Corticosteroid Randomization After Significant Head injury [CRASH], the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries [GUSTO], and PROspective Study of Pravastatin in elderly individuals at risk of vascular disease [PROSPER]). To mimic the RD design, we selected patients above and below a cutoff (e.g., age 75 years) randomized to treatment and control, respectively. Adjusted logistic regression models using restricted cubic splines (RCS) and polynomials and local logistic regression models estimated the odds ratio (OR) for treatment, with 95% confidence intervals (CIs) to indicate precision. In CRASH, treatment increased mortality with OR 1.22 [95% CI 1.06-1.40] in the RCT. The RD estimates were 1.42 (0.94-2.16) and 1.13 (0.90-1.40) with RCS adjustment and local regression, respectively. In GUSTO, treatment reduced mortality (OR 0.83 [0.72-0.95]), with more extreme estimates in the RD analysis (OR 0.57 [0.35; 0.92] and 0.67 [0.51; 0.86]). In PROSPER, similar RCT and RD estimates were found, again with less precision in RD designs. We conclude that the RD design provides similar but substantially less precise treatment effect estimates compared with an RCT, with local regression being the preferred method of analysis. Copyright © 2018 Elsevier Inc. All rights reserved.

Estimation of genetic parameters related to eggshell strength using random regression models.

PubMed

Guo, J; Ma, M; Qu, L; Shen, M; Dou, T; Wang, K

2015-01-01

This study examined the changes in eggshell strength and the genetic parameters related to this trait throughout a hen's laying life using random regression. The data were collected from a crossbred population between 2011 and 2014, where the eggshell strength was determined repeatedly for 2260 hens. Using random regression models (RRMs), several Legendre polynomials were employed to estimate the fixed, direct genetic and permanent environment effects. The residual effects were treated as independently distributed with heterogeneous variance for each test week. The direct genetic variance was included with second-order Legendre polynomials and the permanent environment with third-order Legendre polynomials. The heritability of eggshell strength ranged from 0.26 to 0.43, the repeatability ranged between 0.47 and 0.69, and the estimated genetic correlations between test weeks was high at > 0.67. The first eigenvalue of the genetic covariance matrix accounted for about 97% of the sum of all the eigenvalues. The flexibility and statistical power of RRM suggest that this model could be an effective method to improve eggshell quality and to reduce losses due to cracked eggs in a breeding plan.

An INAR(1) Negative Multinomial Regression Model for Longitudinal Count Data.

ERIC Educational Resources Information Center

Bockenholt, Ulf

1999-01-01

Discusses a regression model for the analysis of longitudinal count data in a panel study by adapting an integer-valued first-order autoregressive (INAR(1)) Poisson process to represent time-dependent correlation between counts. Derives a new negative multinomial distribution by combining INAR(1) representation with a random effects approach.…

Development of a hybrid proximal sensing method for rapid identification of petroleum contaminated soils.

PubMed

Chakraborty, Somsubhra; Weindorf, David C; Li, Bin; Ali Aldabaa, Abdalsamad Abdalsatar; Ghosh, Rakesh Kumar; Paul, Sathi; Nasim Ali, Md

2015-05-01

Using 108 petroleum contaminated soil samples, this pilot study proposed a new analytical approach of combining visible near-infrared diffuse reflectance spectroscopy (VisNIR DRS) and portable X-ray fluorescence spectrometry (PXRF) for rapid and improved quantification of soil petroleum contamination. Results indicated that an advanced fused model where VisNIR DRS spectra-based penalized spline regression (PSR) was used to predict total petroleum hydrocarbon followed by PXRF elemental data-based random forest regression was used to model the PSR residuals, it outperformed (R(2)=0.78, residual prediction deviation (RPD)=2.19) all other models tested, even producing better generalization than using VisNIR DRS alone (RPD's of 1.64, 1.86, and 1.96 for random forest, penalized spline regression, and partial least squares regression, respectively). Additionally, unsupervised principal component analysis using the PXRF+VisNIR DRS system qualitatively separated contaminated soils from control samples. Fusion of PXRF elemental data and VisNIR derivative spectra produced an optimized model for total petroleum hydrocarbon quantification in soils. Copyright © 2015 Elsevier B.V. All rights reserved.

ToxiM: A Toxicity Prediction Tool for Small Molecules Developed Using Machine Learning and Chemoinformatics Approaches.

PubMed

Sharma, Ashok K; Srivastava, Gopal N; Roy, Ankita; Sharma, Vineet K

2017-01-01

The experimental methods for the prediction of molecular toxicity are tedious and time-consuming tasks. Thus, the computational approaches could be used to develop alternative methods for toxicity prediction. We have developed a tool for the prediction of molecular toxicity along with the aqueous solubility and permeability of any molecule/metabolite. Using a comprehensive and curated set of toxin molecules as a training set, the different chemical and structural based features such as descriptors and fingerprints were exploited for feature selection, optimization and development of machine learning based classification and regression models. The compositional differences in the distribution of atoms were apparent between toxins and non-toxins, and hence, the molecular features were used for the classification and regression. On 10-fold cross-validation, the descriptor-based, fingerprint-based and hybrid-based classification models showed similar accuracy (93%) and Matthews's correlation coefficient (0.84). The performances of all the three models were comparable (Matthews's correlation coefficient = 0.84-0.87) on the blind dataset. In addition, the regression-based models using descriptors as input features were also compared and evaluated on the blind dataset. Random forest based regression model for the prediction of solubility performed better ( R 2 = 0.84) than the multi-linear regression (MLR) and partial least square regression (PLSR) models, whereas, the partial least squares based regression model for the prediction of permeability (caco-2) performed better ( R 2 = 0.68) in comparison to the random forest and MLR based regression models. The performance of final classification and regression models was evaluated using the two validation datasets including the known toxins and commonly used constituents of health products, which attests to its accuracy. The ToxiM web server would be a highly useful and reliable tool for the prediction of toxicity, solubility, and permeability of small molecules.

ToxiM: A Toxicity Prediction Tool for Small Molecules Developed Using Machine Learning and Chemoinformatics Approaches

PubMed Central

Sharma, Ashok K.; Srivastava, Gopal N.; Roy, Ankita; Sharma, Vineet K.

2017-01-01

The experimental methods for the prediction of molecular toxicity are tedious and time-consuming tasks. Thus, the computational approaches could be used to develop alternative methods for toxicity prediction. We have developed a tool for the prediction of molecular toxicity along with the aqueous solubility and permeability of any molecule/metabolite. Using a comprehensive and curated set of toxin molecules as a training set, the different chemical and structural based features such as descriptors and fingerprints were exploited for feature selection, optimization and development of machine learning based classification and regression models. The compositional differences in the distribution of atoms were apparent between toxins and non-toxins, and hence, the molecular features were used for the classification and regression. On 10-fold cross-validation, the descriptor-based, fingerprint-based and hybrid-based classification models showed similar accuracy (93%) and Matthews's correlation coefficient (0.84). The performances of all the three models were comparable (Matthews's correlation coefficient = 0.84–0.87) on the blind dataset. In addition, the regression-based models using descriptors as input features were also compared and evaluated on the blind dataset. Random forest based regression model for the prediction of solubility performed better (R2 = 0.84) than the multi-linear regression (MLR) and partial least square regression (PLSR) models, whereas, the partial least squares based regression model for the prediction of permeability (caco-2) performed better (R2 = 0.68) in comparison to the random forest and MLR based regression models. The performance of final classification and regression models was evaluated using the two validation datasets including the known toxins and commonly used constituents of health products, which attests to its accuracy. The ToxiM web server would be a highly useful and reliable tool for the prediction of toxicity, solubility, and permeability of small molecules. PMID:29249969

Comparing spatial regression to random forests for large ...

EPA Pesticide Factsheets

Environmental data may be “large” due to number of records, number of covariates, or both. Random forests has a reputation for good predictive performance when using many covariates, whereas spatial regression, when using reduced rank methods, has a reputation for good predictive performance when using many records. In this study, we compare these two techniques using a data set containing the macroinvertebrate multimetric index (MMI) at 1859 stream sites with over 200 landscape covariates. Our primary goal is predicting MMI at over 1.1 million perennial stream reaches across the USA. For spatial regression modeling, we develop two new methods to accommodate large data: (1) a procedure that estimates optimal Box-Cox transformations to linearize covariate relationships; and (2) a computationally efficient covariate selection routine that takes into account spatial autocorrelation. We show that our new methods lead to cross-validated performance similar to random forests, but that there is an advantage for spatial regression when quantifying the uncertainty of the predictions. Simulations are used to clarify advantages for each method. This research investigates different approaches for modeling and mapping national stream condition. We use MMI data from the EPA's National Rivers and Streams Assessment and predictors from StreamCat (Hill et al., 2015). Previous studies have focused on modeling the MMI condition classes (i.e., good, fair, and po

Random Regression Models Using Legendre Polynomials to Estimate Genetic Parameters for Test-day Milk Protein Yields in Iranian Holstein Dairy Cattle.

PubMed

Naserkheil, Masoumeh; Miraie-Ashtiani, Seyed Reza; Nejati-Javaremi, Ardeshir; Son, Jihyun; Lee, Deukhwan

2016-12-01

The objective of this study was to estimate the genetic parameters of milk protein yields in Iranian Holstein dairy cattle. A total of 1,112,082 test-day milk protein yield records of 167,269 first lactation Holstein cows, calved from 1990 to 2010, were analyzed. Estimates of the variance components, heritability, and genetic correlations for milk protein yields were obtained using a random regression test-day model. Milking times, herd, age of recording, year, and month of recording were included as fixed effects in the model. Additive genetic and permanent environmental random effects for the lactation curve were taken into account by applying orthogonal Legendre polynomials of the fourth order in the model. The lowest and highest additive genetic variances were estimated at the beginning and end of lactation, respectively. Permanent environmental variance was higher at both extremes. Residual variance was lowest at the middle of the lactation and contrarily, heritability increased during this period. Maximum heritability was found during the 12th lactation stage (0.213±0.007). Genetic, permanent, and phenotypic correlations among test-days decreased as the interval between consecutive test-days increased. A relatively large data set was used in this study; therefore, the estimated (co)variance components for random regression coefficients could be used for national genetic evaluation of dairy cattle in Iran.

Random Regression Models Using Legendre Polynomials to Estimate Genetic Parameters for Test-day Milk Protein Yields in Iranian Holstein Dairy Cattle

PubMed Central

Naserkheil, Masoumeh; Miraie-Ashtiani, Seyed Reza; Nejati-Javaremi, Ardeshir; Son, Jihyun; Lee, Deukhwan

2016-01-01

The objective of this study was to estimate the genetic parameters of milk protein yields in Iranian Holstein dairy cattle. A total of 1,112,082 test-day milk protein yield records of 167,269 first lactation Holstein cows, calved from 1990 to 2010, were analyzed. Estimates of the variance components, heritability, and genetic correlations for milk protein yields were obtained using a random regression test-day model. Milking times, herd, age of recording, year, and month of recording were included as fixed effects in the model. Additive genetic and permanent environmental random effects for the lactation curve were taken into account by applying orthogonal Legendre polynomials of the fourth order in the model. The lowest and highest additive genetic variances were estimated at the beginning and end of lactation, respectively. Permanent environmental variance was higher at both extremes. Residual variance was lowest at the middle of the lactation and contrarily, heritability increased during this period. Maximum heritability was found during the 12th lactation stage (0.213±0.007). Genetic, permanent, and phenotypic correlations among test-days decreased as the interval between consecutive test-days increased. A relatively large data set was used in this study; therefore, the estimated (co)variance components for random regression coefficients could be used for national genetic evaluation of dairy cattle in Iran. PMID:26954192

Summer School Effects in a Randomized Field Trial

ERIC Educational Resources Information Center

Zvoch, Keith; Stevens, Joseph J.

2013-01-01

This field-based randomized trial examined the effect of assignment to and participation in summer school for two moderately at-risk samples of struggling readers. Application of multiple regression models to difference scores capturing the change in summer reading fluency revealed that kindergarten students randomly assigned to summer school…

Methods for identifying SNP interactions: a review on variations of Logic Regression, Random Forest and Bayesian logistic regression.

PubMed

Chen, Carla Chia-Ming; Schwender, Holger; Keith, Jonathan; Nunkesser, Robin; Mengersen, Kerrie; Macrossan, Paula

2011-01-01

Due to advancements in computational ability, enhanced technology and a reduction in the price of genotyping, more data are being generated for understanding genetic associations with diseases and disorders. However, with the availability of large data sets comes the inherent challenges of new methods of statistical analysis and modeling. Considering a complex phenotype may be the effect of a combination of multiple loci, various statistical methods have been developed for identifying genetic epistasis effects. Among these methods, logic regression (LR) is an intriguing approach incorporating tree-like structures. Various methods have built on the original LR to improve different aspects of the model. In this study, we review four variations of LR, namely Logic Feature Selection, Monte Carlo Logic Regression, Genetic Programming for Association Studies, and Modified Logic Regression-Gene Expression Programming, and investigate the performance of each method using simulated and real genotype data. We contrast these with another tree-like approach, namely Random Forests, and a Bayesian logistic regression with stochastic search variable selection.

Multivariate random-parameters zero-inflated negative binomial regression model: an application to estimate crash frequencies at intersections.

PubMed

Dong, Chunjiao; Clarke, David B; Yan, Xuedong; Khattak, Asad; Huang, Baoshan

2014-09-01

Crash data are collected through police reports and integrated with road inventory data for further analysis. Integrated police reports and inventory data yield correlated multivariate data for roadway entities (e.g., segments or intersections). Analysis of such data reveals important relationships that can help focus on high-risk situations and coming up with safety countermeasures. To understand relationships between crash frequencies and associated variables, while taking full advantage of the available data, multivariate random-parameters models are appropriate since they can simultaneously consider the correlation among the specific crash types and account for unobserved heterogeneity. However, a key issue that arises with correlated multivariate data is the number of crash-free samples increases, as crash counts have many categories. In this paper, we describe a multivariate random-parameters zero-inflated negative binomial (MRZINB) regression model for jointly modeling crash counts. The full Bayesian method is employed to estimate the model parameters. Crash frequencies at urban signalized intersections in Tennessee are analyzed. The paper investigates the performance of MZINB and MRZINB regression models in establishing the relationship between crash frequencies, pavement conditions, traffic factors, and geometric design features of roadway intersections. Compared to the MZINB model, the MRZINB model identifies additional statistically significant factors and provides better goodness of fit in developing the relationships. The empirical results show that MRZINB model possesses most of the desirable statistical properties in terms of its ability to accommodate unobserved heterogeneity and excess zero counts in correlated data. Notably, in the random-parameters MZINB model, the estimated parameters vary significantly across intersections for different crash types. Copyright © 2014 Elsevier Ltd. All rights reserved.

Simultaneous confidence bands for Cox regression from semiparametric random censorship.

PubMed

Mondal, Shoubhik; Subramanian, Sundarraman

2016-01-01

Cox regression is combined with semiparametric random censorship models to construct simultaneous confidence bands (SCBs) for subject-specific survival curves. Simulation results are presented to compare the performance of the proposed SCBs with the SCBs that are based only on standard Cox. The new SCBs provide correct empirical coverage and are more informative. The proposed SCBs are illustrated with two real examples. An extension to handle missing censoring indicators is also outlined.

Model selection for logistic regression models

NASA Astrophysics Data System (ADS)

Duller, Christine

2012-09-01

Model selection for logistic regression models decides which of some given potential regressors have an effect and hence should be included in the final model. The second interesting question is whether a certain factor is heterogeneous among some subsets, i.e. whether the model should include a random intercept or not. In this paper these questions will be answered with classical as well as with Bayesian methods. The application show some results of recent research projects in medicine and business administration.

Genetic analyses of partial egg production in Japanese quail using multi-trait random regression models.

PubMed

Karami, K; Zerehdaran, S; Barzanooni, B; Lotfi, E

2017-12-01

1. The aim of the present study was to estimate genetic parameters for average egg weight (EW) and egg number (EN) at different ages in Japanese quail using multi-trait random regression (MTRR) models. 2. A total of 8534 records from 900 quail, hatched between 2014 and 2015, were used in the study. Average weekly egg weights and egg numbers were measured from second until sixth week of egg production. 3. Nine random regression models were compared to identify the best order of the Legendre polynomials (LP). The most optimal model was identified by the Bayesian Information Criterion. A model with second order of LP for fixed effects, second order of LP for additive genetic effects and third order of LP for permanent environmental effects (MTRR23) was found to be the best. 4. According to the MTRR23 model, direct heritability for EW increased from 0.26 in the second week to 0.53 in the sixth week of egg production, whereas the ratio of permanent environment to phenotypic variance decreased from 0.48 to 0.1. Direct heritability for EN was low, whereas the ratio of permanent environment to phenotypic variance decreased from 0.57 to 0.15 during the production period. 5. For each trait, estimated genetic correlations among weeks of egg production were high (from 0.85 to 0.98). Genetic correlations between EW and EN were low and negative for the first two weeks, but they were low and positive for the rest of the egg production period. 6. In conclusion, random regression models can be used effectively for analysing egg production traits in Japanese quail. Response to selection for increased egg weight would be higher at older ages because of its higher heritability and such a breeding program would have no negative genetic impact on egg production.

Random regression models using Legendre polynomials or linear splines for test-day milk yield of dairy Gyr (Bos indicus) cattle.

PubMed

Pereira, R J; Bignardi, A B; El Faro, L; Verneque, R S; Vercesi Filho, A E; Albuquerque, L G

2013-01-01

Studies investigating the use of random regression models for genetic evaluation of milk production in Zebu cattle are scarce. In this study, 59,744 test-day milk yield records from 7,810 first lactations of purebred dairy Gyr (Bos indicus) and crossbred (dairy Gyr × Holstein) cows were used to compare random regression models in which additive genetic and permanent environmental effects were modeled using orthogonal Legendre polynomials or linear spline functions. Residual variances were modeled considering 1, 5, or 10 classes of days in milk. Five classes fitted the changes in residual variances over the lactation adequately and were used for model comparison. The model that fitted linear spline functions with 6 knots provided the lowest sum of residual variances across lactation. On the other hand, according to the deviance information criterion (DIC) and bayesian information criterion (BIC), a model using third-order and fourth-order Legendre polynomials for additive genetic and permanent environmental effects, respectively, provided the best fit. However, the high rank correlation (0.998) between this model and that applying third-order Legendre polynomials for additive genetic and permanent environmental effects, indicates that, in practice, the same bulls would be selected by both models. The last model, which is less parameterized, is a parsimonious option for fitting dairy Gyr breed test-day milk yield records. Copyright © 2013 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

A Bayesian ridge regression analysis of congestion's impact on urban expressway safety.

PubMed

Shi, Qi; Abdel-Aty, Mohamed; Lee, Jaeyoung

2016-03-01

With the rapid growth of traffic in urban areas, concerns about congestion and traffic safety have been heightened. This study leveraged both Automatic Vehicle Identification (AVI) system and Microwave Vehicle Detection System (MVDS) installed on an expressway in Central Florida to explore how congestion impacts the crash occurrence in urban areas. Multiple congestion measures from the two systems were developed. To ensure more precise estimates of the congestion's effects, the traffic data were aggregated into peak and non-peak hours. Multicollinearity among traffic parameters was examined. The results showed the presence of multicollinearity especially during peak hours. As a response, ridge regression was introduced to cope with this issue. Poisson models with uncorrelated random effects, correlated random effects, and both correlated random effects and random parameters were constructed within the Bayesian framework. It was proven that correlated random effects could significantly enhance model performance. The random parameters model has similar goodness-of-fit compared with the model with only correlated random effects. However, by accounting for the unobserved heterogeneity, more variables were found to be significantly related to crash frequency. The models indicated that congestion increased crash frequency during peak hours while during non-peak hours it was not a major crash contributing factor. Using the random parameter model, the three congestion measures were compared. It was found that all congestion indicators had similar effects while Congestion Index (CI) derived from MVDS data was a better congestion indicator for safety analysis. Also, analyses showed that the segments with higher congestion intensity could not only increase property damage only (PDO) crashes, but also more severe crashes. In addition, the issues regarding the necessity to incorporate specific congestion indicator for congestion's effects on safety and to take care of the multicollinearity between explanatory variables were also discussed. By including a specific congestion indicator, the model performance significantly improved. When comparing models with and without ridge regression, the magnitude of the coefficients was altered in the existence of multicollinearity. These conclusions suggest that the use of appropriate congestion measure and consideration of multicolilnearity among the variables would improve the models and our understanding about the effects of congestion on traffic safety. Copyright © 2015 Elsevier Ltd. All rights reserved.

Testing the Hypothesis of a Homoscedastic Error Term in Simple, Nonparametric Regression

ERIC Educational Resources Information Center

Wilcox, Rand R.

2006-01-01

Consider the nonparametric regression model Y = m(X)+ [tau](X)[epsilon], where X and [epsilon] are independent random variables, [epsilon] has a median of zero and variance [sigma][squared], [tau] is some unknown function used to model heteroscedasticity, and m(X) is an unknown function reflecting some conditional measure of location associated…

Mortality risk prediction in burn injury: Comparison of logistic regression with machine learning approaches.

PubMed

Stylianou, Neophytos; Akbarov, Artur; Kontopantelis, Evangelos; Buchan, Iain; Dunn, Ken W

2015-08-01

Predicting mortality from burn injury has traditionally employed logistic regression models. Alternative machine learning methods have been introduced in some areas of clinical prediction as the necessary software and computational facilities have become accessible. Here we compare logistic regression and machine learning predictions of mortality from burn. An established logistic mortality model was compared to machine learning methods (artificial neural network, support vector machine, random forests and naïve Bayes) using a population-based (England & Wales) case-cohort registry. Predictive evaluation used: area under the receiver operating characteristic curve; sensitivity; specificity; positive predictive value and Youden's index. All methods had comparable discriminatory abilities, similar sensitivities, specificities and positive predictive values. Although some machine learning methods performed marginally better than logistic regression the differences were seldom statistically significant and clinically insubstantial. Random forests were marginally better for high positive predictive value and reasonable sensitivity. Neural networks yielded slightly better prediction overall. Logistic regression gives an optimal mix of performance and interpretability. The established logistic regression model of burn mortality performs well against more complex alternatives. Clinical prediction with a small set of strong, stable, independent predictors is unlikely to gain much from machine learning outside specialist research contexts. Copyright © 2015 Elsevier Ltd and ISBI. All rights reserved.

False Positives in Multiple Regression: Unanticipated Consequences of Measurement Error in the Predictor Variables

ERIC Educational Resources Information Center

Shear, Benjamin R.; Zumbo, Bruno D.

2013-01-01

Type I error rates in multiple regression, and hence the chance for false positive research findings, can be drastically inflated when multiple regression models are used to analyze data that contain random measurement error. This article shows the potential for inflated Type I error rates in commonly encountered scenarios and provides new…

Comparison of Logistic Regression and Random Forests techniques for shallow landslide susceptibility assessment in Giampilieri (NE Sicily, Italy)

NASA Astrophysics Data System (ADS)

Trigila, Alessandro; Iadanza, Carla; Esposito, Carlo; Scarascia-Mugnozza, Gabriele

2015-11-01

The aim of this work is to define reliable susceptibility models for shallow landslides using Logistic Regression and Random Forests multivariate statistical techniques. The study area, located in North-East Sicily, was hit on October 1st 2009 by a severe rainstorm (225 mm of cumulative rainfall in 7 h) which caused flash floods and more than 1000 landslides. Several small villages, such as Giampilieri, were hit with 31 fatalities, 6 missing persons and damage to buildings and transportation infrastructures. Landslides, mainly types such as earth and debris translational slides evolving into debris flows, were triggered on steep slopes and involved colluvium and regolith materials which cover the underlying metamorphic bedrock. The work has been carried out with the following steps: i) realization of a detailed event landslide inventory map through field surveys coupled with observation of high resolution aerial colour orthophoto; ii) identification of landslide source areas; iii) data preparation of landslide controlling factors and descriptive statistics based on a bivariate method (Frequency Ratio) to get an initial overview on existing relationships between causative factors and shallow landslide source areas; iv) choice of criteria for the selection and sizing of the mapping unit; v) implementation of 5 multivariate statistical susceptibility models based on Logistic Regression and Random Forests techniques and focused on landslide source areas; vi) evaluation of the influence of sample size and type of sampling on results and performance of the models; vii) evaluation of the predictive capabilities of the models using ROC curve, AUC and contingency tables; viii) comparison of model results and obtained susceptibility maps; and ix) analysis of temporal variation of landslide susceptibility related to input parameter changes. Models based on Logistic Regression and Random Forests have demonstrated excellent predictive capabilities. Land use and wildfire variables were found to have a strong control on the occurrence of very rapid shallow landslides.

Random regression models for the prediction of days to weight, ultrasound rib eye area, and ultrasound back fat depth in beef cattle.

PubMed

Speidel, S E; Peel, R K; Crews, D H; Enns, R M

2016-02-01

Genetic evaluation research designed to reduce the required days to a specified end point has received very little attention in pertinent scientific literature, given that its economic importance was first discussed in 1957. There are many production scenarios in today's beef industry, making a prediction for the required number of days to a single end point a suboptimal option. Random regression is an attractive alternative to calculate days to weight (DTW), days to ultrasound back fat (DTUBF), and days to ultrasound rib eye area (DTUREA) genetic predictions that could overcome weaknesses of a single end point prediction. The objective of this study was to develop random regression approaches for the prediction of the DTW, DTUREA, and DTUBF. Data were obtained from the Agriculture and Agri-Food Canada Research Centre, Lethbridge, AB, Canada. Data consisted of records on 1,324 feedlot cattle spanning 1999 to 2007. Individual animals averaged 5.77 observations with weights, ultrasound rib eye area (UREA), ultrasound back fat depth (UBF), and ages ranging from 293 to 863 kg, 73.39 to 129.54 cm, 1.53 to 30.47 mm, and 276 to 519 d, respectively. Random regression models using Legendre polynomials were used to regress age of the individual on weight, UREA, and UBF. Fixed effects in the model included an overall fixed regression of age on end point (weight, UREA, and UBF) nested within breed to account for the mean relationship between age and weight as well as a contemporary group effect consisting of breed of the animal (Angus, Charolais, and Charolais sired), feedlot pen, and year of measure. Likelihood ratio tests were used to determine the appropriate random polynomial order. Use of the quadratic polynomial did not account for any additional genetic variation in days for DTW ( > 0.11), for DTUREA ( > 0.18), and for DTUBF ( > 0.20) when compared with the linear random polynomial. Heritability estimates from the linear random regression for DTW ranged from 0.54 to 0.74, corresponding to end points of 293 and 863 kg, respectively. Heritability for DTUREA ranged from 0.51 to 0.34 and for DTUBF ranged from 0.55 to 0.37. These estimates correspond to UREA end points of 35 and 125 cm and UBF end points of 1.53 and 30 mm, respectively. This range of heritability shows DTW, DTUREA, and DTUBF to be highly heritable and indicates that selection pressure aimed at reducing the number of days to reach a finish weight end point can result in genetic change given sufficient data.

Technology diffusion in hospitals: a log odds random effects regression model.

PubMed

Blank, Jos L T; Valdmanis, Vivian G

2015-01-01

This study identifies the factors that affect the diffusion of hospital innovations. We apply a log odds random effects regression model on hospital micro data. We introduce the concept of clustering innovations and the application of a log odds random effects regression model to describe the diffusion of technologies. We distinguish a number of determinants, such as service, physician, and environmental, financial and organizational characteristics of the 60 Dutch hospitals in our sample. On the basis of this data set on Dutch general hospitals over the period 1995-2002, we conclude that there is a relation between a number of determinants and the diffusion of innovations underlining conclusions from earlier research. Positive effects were found on the basis of the size of the hospitals, competition and a hospital's commitment to innovation. It appears that if a policy is developed to further diffuse innovations, the external effects of demand and market competition need to be examined, which would de facto lead to an efficient use of technology. For the individual hospital, instituting an innovations office appears to be the most prudent course of action. © 2013 The Authors. International Journal of Health Planning and Management published by John Wiley & Sons, Ltd.

Genetic parameters for stayability to consecutive calvings in Zebu cattle.

PubMed

Silva, D O; Santana, M L; Ayres, D R; Menezes, G R O; Silva, L O C; Nobre, P R C; Pereira, R J

2017-12-22

Longer-lived cows tend to be more profitable and the stayability trait is a selection criterion correlated to longevity. An alternative to the traditional approach to evaluate stayability is its definition based on consecutive calvings, whose main advantage is the more accurate evaluation of young bulls. However, no study using this alternative approach has been conducted for Zebu breeds. Therefore, the objective of this study was to compare linear random regression models to fit stayability to consecutive calvings of Guzerá, Nelore and Tabapuã cows and to estimate genetic parameters for this trait in the respective breeds. Data up to the eighth calving were used. The models included the fixed effects of age at first calving and year-season of birth of the cow and the random effects of contemporary group, additive genetic, permanent environmental and residual. Random regressions were modeled by orthogonal Legendre polynomials of order 1 to 4 (2 to 5 coefficients) for contemporary group, additive genetic and permanent environmental effects. Using Deviance Information Criterion as the selection criterion, the model with 4 regression coefficients for each effect was the most adequate for the Nelore and Tabapuã breeds and the model with 5 coefficients is recommended for the Guzerá breed. For Guzerá, heritabilities ranged from 0.05 to 0.08, showing a quadratic trend with a peak between the fourth and sixth calving. For the Nelore and Tabapuã breeds, the estimates ranged from 0.03 to 0.07 and from 0.03 to 0.08, respectively, and increased with increasing calving number. The additive genetic correlations exhibited a similar trend among breeds and were higher for stayability between closer calvings. Even between more distant calvings (second v. eighth), stayability showed a moderate to high genetic correlation, which was 0.77, 0.57 and 0.79 for the Guzerá, Nelore and Tabapuã breeds, respectively. For Guzerá, when the models with 4 or 5 regression coefficients were compared, the rank correlations between predicted breeding values for the intercept were always higher than 0.99, indicating the possibility of practical application of the least parameterized model. In conclusion, the model with 4 random regression coefficients is recommended for the genetic evaluation of stayability to consecutive calvings in Zebu cattle.

Comparison of the Predictive Performance and Interpretability of Random Forest and Linear Models on Benchmark Data Sets.

PubMed

Marchese Robinson, Richard L; Palczewska, Anna; Palczewski, Jan; Kidley, Nathan

2017-08-28

The ability to interpret the predictions made by quantitative structure-activity relationships (QSARs) offers a number of advantages. While QSARs built using nonlinear modeling approaches, such as the popular Random Forest algorithm, might sometimes be more predictive than those built using linear modeling approaches, their predictions have been perceived as difficult to interpret. However, a growing number of approaches have been proposed for interpreting nonlinear QSAR models in general and Random Forest in particular. In the current work, we compare the performance of Random Forest to those of two widely used linear modeling approaches: linear Support Vector Machines (SVMs) (or Support Vector Regression (SVR)) and partial least-squares (PLS). We compare their performance in terms of their predictivity as well as the chemical interpretability of the predictions using novel scoring schemes for assessing heat map images of substructural contributions. We critically assess different approaches for interpreting Random Forest models as well as for obtaining predictions from the forest. We assess the models on a large number of widely employed public-domain benchmark data sets corresponding to regression and binary classification problems of relevance to hit identification and toxicology. We conclude that Random Forest typically yields comparable or possibly better predictive performance than the linear modeling approaches and that its predictions may also be interpreted in a chemically and biologically meaningful way. In contrast to earlier work looking at interpretation of nonlinear QSAR models, we directly compare two methodologically distinct approaches for interpreting Random Forest models. The approaches for interpreting Random Forest assessed in our article were implemented using open-source programs that we have made available to the community. These programs are the rfFC package ( https://r-forge.r-project.org/R/?group_id=1725 ) for the R statistical programming language and the Python program HeatMapWrapper [ https://doi.org/10.5281/zenodo.495163 ] for heat map generation.

Bayesian random-effect model for predicting outcome fraught with heterogeneity--an illustration with episodes of 44 patients with intractable epilepsy.

PubMed

Yen, A M-F; Liou, H-H; Lin, H-L; Chen, T H-H

2006-01-01

The study aimed to develop a predictive model to deal with data fraught with heterogeneity that cannot be explained by sampling variation or measured covariates. The random-effect Poisson regression model was first proposed to deal with over-dispersion for data fraught with heterogeneity after making allowance for measured covariates. Bayesian acyclic graphic model in conjunction with Markov Chain Monte Carlo (MCMC) technique was then applied to estimate the parameters of both relevant covariates and random effect. Predictive distribution was then generated to compare the predicted with the observed for the Bayesian model with and without random effect. Data from repeated measurement of episodes among 44 patients with intractable epilepsy were used as an illustration. The application of Poisson regression without taking heterogeneity into account to epilepsy data yielded a large value of heterogeneity (heterogeneity factor = 17.90, deviance = 1485, degree of freedom (df) = 83). After taking the random effect into account, the value of heterogeneity factor was greatly reduced (heterogeneity factor = 0.52, deviance = 42.5, df = 81). The Pearson chi2 for the comparison between the expected seizure frequencies and the observed ones at two and three months of the model with and without random effect were 34.27 (p = 1.00) and 1799.90 (p < 0.0001), respectively. The Bayesian acyclic model using the MCMC method was demonstrated to have great potential for disease prediction while data show over-dispersion attributed either to correlated property or to subject-to-subject variability.

[Establishment of multiple regression model for virulence factors of Saccharomyces albicans by random amplified polymorphic DNA bands].

PubMed

Liu, Qi; Wu, Youcong; Yuan, Youhua; Bai, Li; Niu, Kun

2011-12-01

To research the relationship between the virulence factors of Saccharomyces albicans (S. albicans) and the random amplified polymorphic DNA (RAPD) bands of them, and establish the regression model by multiple regression analysis. Extracellular phospholipase, secreted proteinase, ability to generate germ tubes and adhere to oral mucosal cells of 92 strains of S. albicans were measured in vitro; RAPD-polymerase chain reaction (RAPD-PCR) was used to get their bands. Multiple regression for virulence factors of S. albicans and RAPD-PCR bands was established. The extracellular phospholipase activity was associated with 4 RAPD bands: 350, 450, 650 and 1 300 bp (P < 0.05); secreted proteinase activity of S. albicans was associated with 2 bands: 350 and 1 200 bp (P < 0.05); the ability of germ tube produce was associated with 2 bands: 400 and 550 bp (P < 0.05). Some RAPD bands will reflect the virulence factors of S. albicans indirectly. These bands would contain some important messages for regulation of S. albicans virulence factors.

Logistic random effects regression models: a comparison of statistical packages for binary and ordinal outcomes.

PubMed

Li, Baoyue; Lingsma, Hester F; Steyerberg, Ewout W; Lesaffre, Emmanuel

2011-05-23

Logistic random effects models are a popular tool to analyze multilevel also called hierarchical data with a binary or ordinal outcome. Here, we aim to compare different statistical software implementations of these models. We used individual patient data from 8509 patients in 231 centers with moderate and severe Traumatic Brain Injury (TBI) enrolled in eight Randomized Controlled Trials (RCTs) and three observational studies. We fitted logistic random effects regression models with the 5-point Glasgow Outcome Scale (GOS) as outcome, both dichotomized as well as ordinal, with center and/or trial as random effects, and as covariates age, motor score, pupil reactivity or trial. We then compared the implementations of frequentist and Bayesian methods to estimate the fixed and random effects. Frequentist approaches included R (lme4), Stata (GLLAMM), SAS (GLIMMIX and NLMIXED), MLwiN ([R]IGLS) and MIXOR, Bayesian approaches included WinBUGS, MLwiN (MCMC), R package MCMCglmm and SAS experimental procedure MCMC.Three data sets (the full data set and two sub-datasets) were analysed using basically two logistic random effects models with either one random effect for the center or two random effects for center and trial. For the ordinal outcome in the full data set also a proportional odds model with a random center effect was fitted. The packages gave similar parameter estimates for both the fixed and random effects and for the binary (and ordinal) models for the main study and when based on a relatively large number of level-1 (patient level) data compared to the number of level-2 (hospital level) data. However, when based on relatively sparse data set, i.e. when the numbers of level-1 and level-2 data units were about the same, the frequentist and Bayesian approaches showed somewhat different results. The software implementations differ considerably in flexibility, computation time, and usability. There are also differences in the availability of additional tools for model evaluation, such as diagnostic plots. The experimental SAS (version 9.2) procedure MCMC appeared to be inefficient. On relatively large data sets, the different software implementations of logistic random effects regression models produced similar results. Thus, for a large data set there seems to be no explicit preference (of course if there is no preference from a philosophical point of view) for either a frequentist or Bayesian approach (if based on vague priors). The choice for a particular implementation may largely depend on the desired flexibility, and the usability of the package. For small data sets the random effects variances are difficult to estimate. In the frequentist approaches the MLE of this variance was often estimated zero with a standard error that is either zero or could not be determined, while for Bayesian methods the estimates could depend on the chosen "non-informative" prior of the variance parameter. The starting value for the variance parameter may be also critical for the convergence of the Markov chain.

Linear Regression with a Randomly Censored Covariate: Application to an Alzheimer's Study.

PubMed

Atem, Folefac D; Qian, Jing; Maye, Jacqueline E; Johnson, Keith A; Betensky, Rebecca A

2017-01-01

The association between maternal age of onset of dementia and amyloid deposition (measured by in vivo positron emission tomography (PET) imaging) in cognitively normal older offspring is of interest. In a regression model for amyloid, special methods are required due to the random right censoring of the covariate of maternal age of onset of dementia. Prior literature has proposed methods to address the problem of censoring due to assay limit of detection, but not random censoring. We propose imputation methods and a survival regression method that do not require parametric assumptions about the distribution of the censored covariate. Existing imputation methods address missing covariates, but not right censored covariates. In simulation studies, we compare these methods to the simple, but inefficient complete case analysis, and to thresholding approaches. We apply the methods to the Alzheimer's study.

Representation of limb kinematics in Purkinje cell simple spike discharge is conserved across multiple tasks

PubMed Central

Hewitt, Angela L.; Popa, Laurentiu S.; Pasalar, Siavash; Hendrix, Claudia M.

2011-01-01

Encoding of movement kinematics in Purkinje cell simple spike discharge has important implications for hypotheses of cerebellar cortical function. Several outstanding questions remain regarding representation of these kinematic signals. It is uncertain whether kinematic encoding occurs in unpredictable, feedback-dependent tasks or kinematic signals are conserved across tasks. Additionally, there is a need to understand the signals encoded in the instantaneous discharge of single cells without averaging across trials or time. To address these questions, this study recorded Purkinje cell firing in monkeys trained to perform a manual random tracking task in addition to circular tracking and center-out reach. Random tracking provides for extensive coverage of kinematic workspaces. Direction and speed errors are significantly greater during random than circular tracking. Cross-correlation analyses comparing hand and target velocity profiles show that hand velocity lags target velocity during random tracking. Correlations between simple spike firing from 120 Purkinje cells and hand position, velocity, and speed were evaluated with linear regression models including a time constant, τ, as a measure of the firing lead/lag relative to the kinematic parameters. Across the population, velocity accounts for the majority of simple spike firing variability (63 ± 30% of Radj2), followed by position (28 ± 24% of Radj2) and speed (11 ± 19% of Radj2). Simple spike firing often leads hand kinematics. Comparison of regression models based on averaged vs. nonaveraged firing and kinematics reveals lower Radj2 values for nonaveraged data; however, regression coefficients and τ values are highly similar. Finally, for most cells, model coefficients generated from random tracking accurately estimate simple spike firing in either circular tracking or center-out reach. These findings imply that the cerebellum controls movement kinematics, consistent with a forward internal model that predicts upcoming limb kinematics. PMID:21795616

Estimating Individual Influences of Behavioral Intentions: An Application of Random-Effects Modeling to the Theory of Reasoned Action.

ERIC Educational Resources Information Center

Hedeker, Donald; And Others

1996-01-01

Methods are proposed and described for estimating the degree to which relations among variables vary at the individual level. As an example, M. Fishbein and I. Ajzen's theory of reasoned action is examined. This article illustrates the use of empirical Bayes methods based on a random-effects regression model to estimate individual influences…

The creation and evaluation of a model to simulate the probability of conception in seasonal-calving pasture-based dairy heifers.

PubMed

Fenlon, Caroline; O'Grady, Luke; Butler, Stephen; Doherty, Michael L; Dunnion, John

2017-01-01

Herd fertility in pasture-based dairy farms is a key driver of farm economics. Models for predicting nulliparous reproductive outcomes are rare, but age, genetics, weight, and BCS have been identified as factors influencing heifer conception. The aim of this study was to create a simulation model of heifer conception to service with thorough evaluation. Artificial Insemination service records from two research herds and ten commercial herds were provided to build and evaluate the models. All were managed as spring-calving pasture-based systems. The factors studied were related to age, genetics, and time of service. The data were split into training and testing sets and bootstrapping was used to train the models. Logistic regression (with and without random effects) and generalised additive modelling were selected as the model-building techniques. Two types of evaluation were used to test the predictive ability of the models: discrimination and calibration. Discrimination, which includes sensitivity, specificity, accuracy and ROC analysis, measures a model's ability to distinguish between positive and negative outcomes. Calibration measures the accuracy of the predicted probabilities with the Hosmer-Lemeshow goodness-of-fit, calibration plot and calibration error. After data cleaning and the removal of services with missing values, 1396 services remained to train the models and 597 were left for testing. Age, breed, genetic predicted transmitting ability for calving interval, month and year were significant in the multivariate models. The regression models also included an interaction between age and month. Year within herd was a random effect in the mixed regression model. Overall prediction accuracy was between 77.1% and 78.9%. All three models had very high sensitivity, but low specificity. The two regression models were very well-calibrated. The mean absolute calibration errors were all below 4%. Because the models were not adept at identifying unsuccessful services, they are not suggested for use in predicting the outcome of individual heifer services. Instead, they are useful for the comparison of services with different covariate values or as sub-models in whole-farm simulations. The mixed regression model was identified as the best model for prediction, as the random effects can be ignored and the other variables can be easily obtained or simulated.

Statistical-learning strategies generate only modestly performing predictive models for urinary symptoms following external beam radiotherapy of the prostate: A comparison of conventional and machine-learning methods

DOE Office of Scientific and Technical Information (OSTI.GOV)

Yahya, Noorazrul, E-mail: noorazrul.yahya@research.uwa.edu.au; Ebert, Martin A.; Bulsara, Max

Purpose: Given the paucity of available data concerning radiotherapy-induced urinary toxicity, it is important to ensure derivation of the most robust models with superior predictive performance. This work explores multiple statistical-learning strategies for prediction of urinary symptoms following external beam radiotherapy of the prostate. Methods: The performance of logistic regression, elastic-net, support-vector machine, random forest, neural network, and multivariate adaptive regression splines (MARS) to predict urinary symptoms was analyzed using data from 754 participants accrued by TROG03.04-RADAR. Predictive features included dose-surface data, comorbidities, and medication-intake. Four symptoms were analyzed: dysuria, haematuria, incontinence, and frequency, each with three definitions (grade ≥more » 1, grade ≥ 2 and longitudinal) with event rate between 2.3% and 76.1%. Repeated cross-validations producing matched models were implemented. A synthetic minority oversampling technique was utilized in endpoints with rare events. Parameter optimization was performed on the training data. Area under the receiver operating characteristic curve (AUROC) was used to compare performance using sample size to detect differences of ≥0.05 at the 95% confidence level. Results: Logistic regression, elastic-net, random forest, MARS, and support-vector machine were the highest-performing statistical-learning strategies in 3, 3, 3, 2, and 1 endpoints, respectively. Logistic regression, MARS, elastic-net, random forest, neural network, and support-vector machine were the best, or were not significantly worse than the best, in 7, 7, 5, 5, 3, and 1 endpoints. The best-performing statistical model was for dysuria grade ≥ 1 with AUROC ± standard deviation of 0.649 ± 0.074 using MARS. For longitudinal frequency and dysuria grade ≥ 1, all strategies produced AUROC>0.6 while all haematuria endpoints and longitudinal incontinence models produced AUROC<0.6. Conclusions: Logistic regression and MARS were most likely to be the best-performing strategy for the prediction of urinary symptoms with elastic-net and random forest producing competitive results. The predictive power of the models was modest and endpoint-dependent. New features, including spatial dose maps, may be necessary to achieve better models.« less

Summer and winter habitat suitability of Marco Polo argali in southeastern Tajikistan: A modeling approach.

PubMed

Salas, Eric Ariel L; Valdez, Raul; Michel, Stefan

2017-11-01

We modeled summer and winter habitat suitability of Marco Polo argali in the Pamir Mountains in southeastern Tajikistan using these statistical algorithms: Generalized Linear Model, Random Forest, Boosted Regression Tree, Maxent, and Multivariate Adaptive Regression Splines. Using sheep occurrence data collected from 2009 to 2015 and a set of selected habitat predictors, we produced summer and winter habitat suitability maps and determined the important habitat suitability predictors for both seasons. Our results demonstrated that argali selected proximity to riparian areas and greenness as the two most relevant variables for summer, and the degree of slope (gentler slopes between 0° to 20°) and Landsat temperature band for winter. The terrain roughness was also among the most important variables in summer and winter models. Aspect was only significant for winter habitat, with argali preferring south-facing mountain slopes. We evaluated various measures of model performance such as the Area Under the Curve (AUC) and the True Skill Statistic (TSS). Comparing the five algorithms, the AUC scored highest for Boosted Regression Tree in summer (AUC = 0.94) and winter model runs (AUC = 0.94). In contrast, Random Forest underperformed in both model runs.

Impact of Contextual Factors on Prostate Cancer Risk and Outcomes

DTIC Science & Technology

2013-07-01

framework with random effects (“frailty models”) while the case-control analyses (Aim 4) will use multilevel unconditional logistic regression models...contextual-level SES on prostate cancer risk within racial/ethnic groups. The survival analyses (Aims 1-3) will utilize a proportional hazards regression

Logistic regression for dichotomized counts.

PubMed

Preisser, John S; Das, Kalyan; Benecha, Habtamu; Stamm, John W

2016-12-01

Sometimes there is interest in a dichotomized outcome indicating whether a count variable is positive or zero. Under this scenario, the application of ordinary logistic regression may result in efficiency loss, which is quantifiable under an assumed model for the counts. In such situations, a shared-parameter hurdle model is investigated for more efficient estimation of regression parameters relating to overall effects of covariates on the dichotomous outcome, while handling count data with many zeroes. One model part provides a logistic regression containing marginal log odds ratio effects of primary interest, while an ancillary model part describes the mean count of a Poisson or negative binomial process in terms of nuisance regression parameters. Asymptotic efficiency of the logistic model parameter estimators of the two-part models is evaluated with respect to ordinary logistic regression. Simulations are used to assess the properties of the models with respect to power and Type I error, the latter investigated under both misspecified and correctly specified models. The methods are applied to data from a randomized clinical trial of three toothpaste formulations to prevent incident dental caries in a large population of Scottish schoolchildren. © The Author(s) 2014.

Regression dilution bias: tools for correction methods and sample size calculation.

PubMed

Berglund, Lars

2012-08-01

Random errors in measurement of a risk factor will introduce downward bias of an estimated association to a disease or a disease marker. This phenomenon is called regression dilution bias. A bias correction may be made with data from a validity study or a reliability study. In this article we give a non-technical description of designs of reliability studies with emphasis on selection of individuals for a repeated measurement, assumptions of measurement error models, and correction methods for the slope in a simple linear regression model where the dependent variable is a continuous variable. Also, we describe situations where correction for regression dilution bias is not appropriate. The methods are illustrated with the association between insulin sensitivity measured with the euglycaemic insulin clamp technique and fasting insulin, where measurement of the latter variable carries noticeable random error. We provide software tools for estimation of a corrected slope in a simple linear regression model assuming data for a continuous dependent variable and a continuous risk factor from a main study and an additional measurement of the risk factor in a reliability study. Also, we supply programs for estimation of the number of individuals needed in the reliability study and for choice of its design. Our conclusion is that correction for regression dilution bias is seldom applied in epidemiological studies. This may cause important effects of risk factors with large measurement errors to be neglected.

Heterogeneous Suppression of Sequential Effects in Random Sequence Generation, but Not in Operant Learning.

PubMed

Shteingart, Hanan; Loewenstein, Yonatan

2016-01-01

There is a long history of experiments in which participants are instructed to generate a long sequence of binary random numbers. The scope of this line of research has shifted over the years from identifying the basic psychological principles and/or the heuristics that lead to deviations from randomness, to one of predicting future choices. In this paper, we used generalized linear regression and the framework of Reinforcement Learning in order to address both points. In particular, we used logistic regression analysis in order to characterize the temporal sequence of participants' choices. Surprisingly, a population analysis indicated that the contribution of the most recent trial has only a weak effect on behavior, compared to more preceding trials, a result that seems irreconcilable with standard sequential effects that decay monotonously with the delay. However, when considering each participant separately, we found that the magnitudes of the sequential effect are a monotonous decreasing function of the delay, yet these individual sequential effects are largely averaged out in a population analysis because of heterogeneity. The substantial behavioral heterogeneity in this task is further demonstrated quantitatively by considering the predictive power of the model. We show that a heterogeneous model of sequential dependencies captures the structure available in random sequence generation. Finally, we show that the results of the logistic regression analysis can be interpreted in the framework of reinforcement learning, allowing us to compare the sequential effects in the random sequence generation task to those in an operant learning task. We show that in contrast to the random sequence generation task, sequential effects in operant learning are far more homogenous across the population. These results suggest that in the random sequence generation task, different participants adopt different cognitive strategies to suppress sequential dependencies when generating the "random" sequences.

Simple to complex modeling of breathing volume using a motion sensor.

PubMed

John, Dinesh; Staudenmayer, John; Freedson, Patty

2013-06-01

To compare simple and complex modeling techniques to estimate categories of low, medium, and high ventilation (VE) from ActiGraph™ activity counts. Vertical axis ActiGraph™ GT1M activity counts, oxygen consumption and VE were measured during treadmill walking and running, sports, household chores and labor-intensive employment activities. Categories of low (<19.3 l/min), medium (19.3 to 35.4 l/min) and high (>35.4 l/min) VEs were derived from activity intensity classifications (light <2.9 METs, moderate 3.0 to 5.9 METs and vigorous >6.0 METs). We examined the accuracy of two simple techniques (multiple regression and activity count cut-point analyses) and one complex (random forest technique) modeling technique in predicting VE from activity counts. Prediction accuracy of the complex random forest technique was marginally better than the simple multiple regression method. Both techniques accurately predicted VE categories almost 80% of the time. The multiple regression and random forest techniques were more accurate (85 to 88%) in predicting medium VE. Both techniques predicted the high VE (70 to 73%) with greater accuracy than low VE (57 to 60%). Actigraph™ cut-points for light, medium and high VEs were <1381, 1381 to 3660 and >3660 cpm. There were minor differences in prediction accuracy between the multiple regression and the random forest technique. This study provides methods to objectively estimate VE categories using activity monitors that can easily be deployed in the field. Objective estimates of VE should provide a better understanding of the dose-response relationship between internal exposure to pollutants and disease. Copyright © 2013 Elsevier B.V. All rights reserved.

pLARmEB: integration of least angle regression with empirical Bayes for multilocus genome-wide association studies.

PubMed

Zhang, J; Feng, J-Y; Ni, Y-L; Wen, Y-J; Niu, Y; Tamba, C L; Yue, C; Song, Q; Zhang, Y-M

2017-06-01

Multilocus genome-wide association studies (GWAS) have become the state-of-the-art procedure to identify quantitative trait nucleotides (QTNs) associated with complex traits. However, implementation of multilocus model in GWAS is still difficult. In this study, we integrated least angle regression with empirical Bayes to perform multilocus GWAS under polygenic background control. We used an algorithm of model transformation that whitened the covariance matrix of the polygenic matrix K and environmental noise. Markers on one chromosome were included simultaneously in a multilocus model and least angle regression was used to select the most potentially associated single-nucleotide polymorphisms (SNPs), whereas the markers on the other chromosomes were used to calculate kinship matrix as polygenic background control. The selected SNPs in multilocus model were further detected for their association with the trait by empirical Bayes and likelihood ratio test. We herein refer to this method as the pLARmEB (polygenic-background-control-based least angle regression plus empirical Bayes). Results from simulation studies showed that pLARmEB was more powerful in QTN detection and more accurate in QTN effect estimation, had less false positive rate and required less computing time than Bayesian hierarchical generalized linear model, efficient mixed model association (EMMA) and least angle regression plus empirical Bayes. pLARmEB, multilocus random-SNP-effect mixed linear model and fast multilocus random-SNP-effect EMMA methods had almost equal power of QTN detection in simulation experiments. However, only pLARmEB identified 48 previously reported genes for 7 flowering time-related traits in Arabidopsis thaliana.

Mathematical models application for mapping soils spatial distribution on the example of the farm from the North of Udmurt Republic of Russia

NASA Astrophysics Data System (ADS)

Dokuchaev, P. M.; Meshalkina, J. L.; Yaroslavtsev, A. M.

2018-01-01

Comparative analysis of soils geospatial modeling using multinomial logistic regression, decision trees, random forest, regression trees and support vector machines algorithms was conducted. The visual interpretation of the digital maps obtained and their comparison with the existing map, as well as the quantitative assessment of the individual soil groups detection overall accuracy and of the models kappa showed that multiple logistic regression, support vector method, and random forest models application with spatial prediction of the conditional soil groups distribution can be reliably used for mapping of the study area. It has shown the most accurate detection for sod-podzolics soils (Phaeozems Albic) lightly eroded and moderately eroded soils. In second place, according to the mean overall accuracy of the prediction, there are sod-podzolics soils - non-eroded and warp one, as well as sod-gley soils (Umbrisols Gleyic) and alluvial soils (Fluvisols Dystric, Umbric). Heavy eroded sod-podzolics and gray forest soils (Phaeozems Albic) were detected by methods of automatic classification worst of all.

Hierarchical Bayesian spatial models for predicting multiple forest variables using waveform LiDAR, hyperspectral imagery, and large inventory datasets

USGS Publications Warehouse

Finley, Andrew O.; Banerjee, Sudipto; Cook, Bruce D.; Bradford, John B.

2013-01-01

In this paper we detail a multivariate spatial regression model that couples LiDAR, hyperspectral and forest inventory data to predict forest outcome variables at a high spatial resolution. The proposed model is used to analyze forest inventory data collected on the US Forest Service Penobscot Experimental Forest (PEF), ME, USA. In addition to helping meet the regression model's assumptions, results from the PEF analysis suggest that the addition of multivariate spatial random effects improves model fit and predictive ability, compared with two commonly applied modeling approaches. This improvement results from explicitly modeling the covariation among forest outcome variables and spatial dependence among observations through the random effects. Direct application of such multivariate models to even moderately large datasets is often computationally infeasible because of cubic order matrix algorithms involved in estimation. We apply a spatial dimension reduction technique to help overcome this computational hurdle without sacrificing richness in modeling.

Solving a mixture of many random linear equations by tensor decomposition and alternating minimization.

DOT National Transportation Integrated Search

2016-09-01

We consider the problem of solving mixed random linear equations with k components. This is the noiseless setting of mixed linear regression. The goal is to estimate multiple linear models from mixed samples in the case where the labels (which sample...

Random forests and stochastic gradient boosting for predicting tree canopy cover: Comparing tuning processes and model performance

Treesearch

E. Freeman; G. Moisen; J. Coulston; B. Wilson

2014-01-01

Random forests (RF) and stochastic gradient boosting (SGB), both involving an ensemble of classification and regression trees, are compared for modeling tree canopy cover for the 2011 National Land Cover Database (NLCD). The objectives of this study were twofold. First, sensitivity of RF and SGB to choices in tuning parameters was explored. Second, performance of the...

Comparison of modeling methods to predict the spatial distribution of deep-sea coral and sponge in the Gulf of Alaska

NASA Astrophysics Data System (ADS)

Rooper, Christopher N.; Zimmermann, Mark; Prescott, Megan M.

2017-08-01

Deep-sea coral and sponge ecosystems are widespread throughout most of Alaska's marine waters, and are associated with many different species of fishes and invertebrates. These ecosystems are vulnerable to the effects of commercial fishing activities and climate change. We compared four commonly used species distribution models (general linear models, generalized additive models, boosted regression trees and random forest models) and an ensemble model to predict the presence or absence and abundance of six groups of benthic invertebrate taxa in the Gulf of Alaska. All four model types performed adequately on training data for predicting presence and absence, with regression forest models having the best overall performance measured by the area under the receiver-operating-curve (AUC). The models also performed well on the test data for presence and absence with average AUCs ranging from 0.66 to 0.82. For the test data, ensemble models performed the best. For abundance data, there was an obvious demarcation in performance between the two regression-based methods (general linear models and generalized additive models), and the tree-based models. The boosted regression tree and random forest models out-performed the other models by a wide margin on both the training and testing data. However, there was a significant drop-off in performance for all models of invertebrate abundance ( 50%) when moving from the training data to the testing data. Ensemble model performance was between the tree-based and regression-based methods. The maps of predictions from the models for both presence and abundance agreed very well across model types, with an increase in variability in predictions for the abundance data. We conclude that where data conforms well to the modeled distribution (such as the presence-absence data and binomial distribution in this study), the four types of models will provide similar results, although the regression-type models may be more consistent with biological theory. For data with highly zero-inflated distributions and non-normal distributions such as the abundance data from this study, the tree-based methods performed better. Ensemble models that averaged predictions across the four model types, performed better than the GLM or GAM models but slightly poorer than the tree-based methods, suggesting ensemble models might be more robust to overfitting than tree methods, while mitigating some of the disadvantages in predictive performance of regression methods.

Modeling nitrate at domestic and public-supply well depths in the Central Valley, California

USGS Publications Warehouse

Nolan, Bernard T.; Gronberg, JoAnn M.; Faunt, Claudia C.; Eberts, Sandra M.; Belitz, Ken

2014-01-01

Aquifer vulnerability models were developed to map groundwater nitrate concentration at domestic and public-supply well depths in the Central Valley, California. We compared three modeling methods for ability to predict nitrate concentration >4 mg/L: logistic regression (LR), random forest classification (RFC), and random forest regression (RFR). All three models indicated processes of nitrogen fertilizer input at the land surface, transmission through coarse-textured, well-drained soils, and transport in the aquifer to the well screen. The total percent correct predictions were similar among the three models (69–82%), but RFR had greater sensitivity (84% for shallow wells and 51% for deep wells). The results suggest that RFR can better identify areas with high nitrate concentration but that LR and RFC may better describe bulk conditions in the aquifer. A unique aspect of the modeling approach was inclusion of outputs from previous, physically based hydrologic and textural models as predictor variables, which were important to the models. Vertical water fluxes in the aquifer and percent coarse material above the well screen were ranked moderately high-to-high in the RFR models, and the average vertical water flux during the irrigation season was highly significant (p < 0.0001) in logistic regression.

GIS-based groundwater potential mapping using boosted regression tree, classification and regression tree, and random forest machine learning models in Iran.

PubMed

Naghibi, Seyed Amir; Pourghasemi, Hamid Reza; Dixon, Barnali

2016-01-01

Groundwater is considered one of the most valuable fresh water resources. The main objective of this study was to produce groundwater spring potential maps in the Koohrang Watershed, Chaharmahal-e-Bakhtiari Province, Iran, using three machine learning models: boosted regression tree (BRT), classification and regression tree (CART), and random forest (RF). Thirteen hydrological-geological-physiographical (HGP) factors that influence locations of springs were considered in this research. These factors include slope degree, slope aspect, altitude, topographic wetness index (TWI), slope length (LS), plan curvature, profile curvature, distance to rivers, distance to faults, lithology, land use, drainage density, and fault density. Subsequently, groundwater spring potential was modeled and mapped using CART, RF, and BRT algorithms. The predicted results from the three models were validated using the receiver operating characteristics curve (ROC). From 864 springs identified, 605 (≈70 %) locations were used for the spring potential mapping, while the remaining 259 (≈30 %) springs were used for the model validation. The area under the curve (AUC) for the BRT model was calculated as 0.8103 and for CART and RF the AUC were 0.7870 and 0.7119, respectively. Therefore, it was concluded that the BRT model produced the best prediction results while predicting locations of springs followed by CART and RF models, respectively. Geospatially integrated BRT, CART, and RF methods proved to be useful in generating the spring potential map (SPM) with reasonable accuracy.

A Tutorial on Multilevel Survival Analysis: Methods, Models and Applications

PubMed Central

Austin, Peter C.

2017-01-01

Summary Data that have a multilevel structure occur frequently across a range of disciplines, including epidemiology, health services research, public health, education and sociology. We describe three families of regression models for the analysis of multilevel survival data. First, Cox proportional hazards models with mixed effects incorporate cluster-specific random effects that modify the baseline hazard function. Second, piecewise exponential survival models partition the duration of follow-up into mutually exclusive intervals and fit a model that assumes that the hazard function is constant within each interval. This is equivalent to a Poisson regression model that incorporates the duration of exposure within each interval. By incorporating cluster-specific random effects, generalised linear mixed models can be used to analyse these data. Third, after partitioning the duration of follow-up into mutually exclusive intervals, one can use discrete time survival models that use a complementary log–log generalised linear model to model the occurrence of the outcome of interest within each interval. Random effects can be incorporated to account for within-cluster homogeneity in outcomes. We illustrate the application of these methods using data consisting of patients hospitalised with a heart attack. We illustrate the application of these methods using three statistical programming languages (R, SAS and Stata). PMID:29307954

A Tutorial on Multilevel Survival Analysis: Methods, Models and Applications.

PubMed

Austin, Peter C

2017-08-01

Data that have a multilevel structure occur frequently across a range of disciplines, including epidemiology, health services research, public health, education and sociology. We describe three families of regression models for the analysis of multilevel survival data. First, Cox proportional hazards models with mixed effects incorporate cluster-specific random effects that modify the baseline hazard function. Second, piecewise exponential survival models partition the duration of follow-up into mutually exclusive intervals and fit a model that assumes that the hazard function is constant within each interval. This is equivalent to a Poisson regression model that incorporates the duration of exposure within each interval. By incorporating cluster-specific random effects, generalised linear mixed models can be used to analyse these data. Third, after partitioning the duration of follow-up into mutually exclusive intervals, one can use discrete time survival models that use a complementary log-log generalised linear model to model the occurrence of the outcome of interest within each interval. Random effects can be incorporated to account for within-cluster homogeneity in outcomes. We illustrate the application of these methods using data consisting of patients hospitalised with a heart attack. We illustrate the application of these methods using three statistical programming languages (R, SAS and Stata).

Logistic quantile regression provides improved estimates for bounded avian counts: A case study of California Spotted Owl fledgling production

USGS Publications Warehouse

Cade, Brian S.; Noon, Barry R.; Scherer, Rick D.; Keane, John J.

2017-01-01

Counts of avian fledglings, nestlings, or clutch size that are bounded below by zero and above by some small integer form a discrete random variable distribution that is not approximated well by conventional parametric count distributions such as the Poisson or negative binomial. We developed a logistic quantile regression model to provide estimates of the empirical conditional distribution of a bounded discrete random variable. The logistic quantile regression model requires that counts are randomly jittered to a continuous random variable, logit transformed to bound them between specified lower and upper values, then estimated in conventional linear quantile regression, repeating the 3 steps and averaging estimates. Back-transformation to the original discrete scale relies on the fact that quantiles are equivariant to monotonic transformations. We demonstrate this statistical procedure by modeling 20 years of California Spotted Owl fledgling production (0−3 per territory) on the Lassen National Forest, California, USA, as related to climate, demographic, and landscape habitat characteristics at territories. Spotted Owl fledgling counts increased nonlinearly with decreasing precipitation in the early nesting period, in the winter prior to nesting, and in the prior growing season; with increasing minimum temperatures in the early nesting period; with adult compared to subadult parents; when there was no fledgling production in the prior year; and when percentage of the landscape surrounding nesting sites (202 ha) with trees ≥25 m height increased. Changes in production were primarily driven by changes in the proportion of territories with 2 or 3 fledglings. Average variances of the discrete cumulative distributions of the estimated fledgling counts indicated that temporal changes in climate and parent age class explained 18% of the annual variance in owl fledgling production, which was 34% of the total variance. Prior fledgling production explained as much of the variance in the fledgling counts as climate, parent age class, and landscape habitat predictors. Our logistic quantile regression model can be used for any discrete response variables with fixed upper and lower bounds.

Extension of the Haseman-Elston regression model to longitudinal data.

PubMed

Won, Sungho; Elston, Robert C; Park, Taesung

2006-01-01

We propose an extension to longitudinal data of the Haseman and Elston regression method for linkage analysis. The proposed model is a mixed model having several random effects. As response variable, we investigate the sibship sample mean corrected cross-product (smHE) and the BLUP-mean corrected cross product (pmHE), comparing them with the original squared difference (oHE), the overall mean corrected cross-product (rHE), and the weighted average of the squared difference and the squared mean-corrected sum (wHE). The proposed model allows for the correlation structure of longitudinal data. Also, the model can test for gene x time interaction to discover genetic variation over time. The model was applied in an analysis of the Genetic Analysis Workshop 13 (GAW13) simulated dataset for a quantitative trait simulating systolic blood pressure. Independence models did not preserve the test sizes, while the mixed models with both family and sibpair random effects tended to preserve size well. Copyright 2006 S. Karger AG, Basel.

Estimating current and future streamflow characteristics at ungaged sites, central and eastern Montana, with application to evaluating effects of climate change on fish populations

USGS Publications Warehouse

Sando, Roy; Chase, Katherine J.

2017-03-23

A common statistical procedure for estimating streamflow statistics at ungaged locations is to develop a relational model between streamflow and drainage basin characteristics at gaged locations using least squares regression analysis; however, least squares regression methods are parametric and make constraining assumptions about the data distribution. The random forest regression method provides an alternative nonparametric method for estimating streamflow characteristics at ungaged sites and requires that the data meet fewer statistical conditions than least squares regression methods.Random forest regression analysis was used to develop predictive models for 89 streamflow characteristics using Precipitation-Runoff Modeling System simulated streamflow data and drainage basin characteristics at 179 sites in central and eastern Montana. The predictive models were developed from streamflow data simulated for current (baseline, water years 1982–99) conditions and three future periods (water years 2021–38, 2046–63, and 2071–88) under three different climate-change scenarios. These predictive models were then used to predict streamflow characteristics for baseline conditions and three future periods at 1,707 fish sampling sites in central and eastern Montana. The average root mean square error for all predictive models was about 50 percent. When streamflow predictions at 23 fish sampling sites were compared to nearby locations with simulated data, the mean relative percent difference was about 43 percent. When predictions were compared to streamflow data recorded at 21 U.S. Geological Survey streamflow-gaging stations outside of the calibration basins, the average mean absolute percent error was about 73 percent.

Neither fixed nor random: weighted least squares meta-regression.

PubMed

Stanley, T D; Doucouliagos, Hristos

2017-03-01

Our study revisits and challenges two core conventional meta-regression estimators: the prevalent use of 'mixed-effects' or random-effects meta-regression analysis and the correction of standard errors that defines fixed-effects meta-regression analysis (FE-MRA). We show how and explain why an unrestricted weighted least squares MRA (WLS-MRA) estimator is superior to conventional random-effects (or mixed-effects) meta-regression when there is publication (or small-sample) bias that is as good as FE-MRA in all cases and better than fixed effects in most practical applications. Simulations and statistical theory show that WLS-MRA provides satisfactory estimates of meta-regression coefficients that are practically equivalent to mixed effects or random effects when there is no publication bias. When there is publication selection bias, WLS-MRA always has smaller bias than mixed effects or random effects. In practical applications, an unrestricted WLS meta-regression is likely to give practically equivalent or superior estimates to fixed-effects, random-effects, and mixed-effects meta-regression approaches. However, random-effects meta-regression remains viable and perhaps somewhat preferable if selection for statistical significance (publication bias) can be ruled out and when random, additive normal heterogeneity is known to directly affect the 'true' regression coefficient. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

Calibrating random forests for probability estimation.

PubMed

Dankowski, Theresa; Ziegler, Andreas

2016-09-30

Probabilities can be consistently estimated using random forests. It is, however, unclear how random forests should be updated to make predictions for other centers or at different time points. In this work, we present two approaches for updating random forests for probability estimation. The first method has been proposed by Elkan and may be used for updating any machine learning approach yielding consistent probabilities, so-called probability machines. The second approach is a new strategy specifically developed for random forests. Using the terminal nodes, which represent conditional probabilities, the random forest is first translated to logistic regression models. These are, in turn, used for re-calibration. The two updating strategies were compared in a simulation study and are illustrated with data from the German Stroke Study Collaboration. In most simulation scenarios, both methods led to similar improvements. In the simulation scenario in which the stricter assumptions of Elkan's method were not met, the logistic regression-based re-calibration approach for random forests outperformed Elkan's method. It also performed better on the stroke data than Elkan's method. The strength of Elkan's method is its general applicability to any probability machine. However, if the strict assumptions underlying this approach are not met, the logistic regression-based approach is preferable for updating random forests for probability estimation. © 2016 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd. © 2016 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

Can Predictive Modeling Identify Head and Neck Oncology Patients at Risk for Readmission?

PubMed

Manning, Amy M; Casper, Keith A; Peter, Kay St; Wilson, Keith M; Mark, Jonathan R; Collar, Ryan M

2018-05-01

Objective Unplanned readmission within 30 days is a contributor to health care costs in the United States. The use of predictive modeling during hospitalization to identify patients at risk for readmission offers a novel approach to quality improvement and cost reduction. Study Design Two-phase study including retrospective analysis of prospectively collected data followed by prospective longitudinal study. Setting Tertiary academic medical center. Subjects and Methods Prospectively collected data for patients undergoing surgical treatment for head and neck cancer from January 2013 to January 2015 were used to build predictive models for readmission within 30 days of discharge using logistic regression, classification and regression tree (CART) analysis, and random forests. One model (logistic regression) was then placed prospectively into the discharge workflow from March 2016 to May 2016 to determine the model's ability to predict which patients would be readmitted within 30 days. Results In total, 174 admissions had descriptive data. Thirty-two were excluded due to incomplete data. Logistic regression, CART, and random forest predictive models were constructed using the remaining 142 admissions. When applied to 106 consecutive prospective head and neck oncology patients at the time of discharge, the logistic regression model predicted readmissions with a specificity of 94%, a sensitivity of 47%, a negative predictive value of 90%, and a positive predictive value of 62% (odds ratio, 14.9; 95% confidence interval, 4.02-55.45). Conclusion Prospectively collected head and neck cancer databases can be used to develop predictive models that can accurately predict which patients will be readmitted. This offers valuable support for quality improvement initiatives and readmission-related cost reduction in head and neck cancer care.

Prediction of Baseflow Index of Catchments using Machine Learning Algorithms

NASA Astrophysics Data System (ADS)

Yadav, B.; Hatfield, K.

2017-12-01

We present the results of eight machine learning techniques for predicting the baseflow index (BFI) of ungauged basins using a surrogate of catchment scale climate and physiographic data. The tested algorithms include ordinary least squares, ridge regression, least absolute shrinkage and selection operator (lasso), elasticnet, support vector machine, gradient boosted regression trees, random forests, and extremely randomized trees. Our work seeks to identify the dominant controls of BFI that can be readily obtained from ancillary geospatial databases and remote sensing measurements, such that the developed techniques can be extended to ungauged catchments. More than 800 gauged catchments spanning the continental United States were selected to develop the general methodology. The BFI calculation was based on the baseflow separated from daily streamflow hydrograph using HYSEP filter. The surrogate catchment attributes were compiled from multiple sources including digital elevation model, soil, landuse, climate data, other publicly available ancillary and geospatial data. 80% catchments were used to train the ML algorithms, and the remaining 20% of the catchments were used as an independent test set to measure the generalization performance of fitted models. A k-fold cross-validation using exhaustive grid search was used to fit the hyperparameters of each model. Initial model development was based on 19 independent variables, but after variable selection and feature ranking, we generated revised sparse models of BFI prediction that are based on only six catchment attributes. These key predictive variables selected after the careful evaluation of bias-variance tradeoff include average catchment elevation, slope, fraction of sand, permeability, temperature, and precipitation. The most promising algorithms exceeding an accuracy score (r-square) of 0.7 on test data include support vector machine, gradient boosted regression trees, random forests, and extremely randomized trees. Considering both the accuracy and the computational complexity of these algorithms, we identify the extremely randomized trees as the best performing algorithm for BFI prediction in ungauged basins.

Predicting recycling behaviour: Comparison of a linear regression model and a fuzzy logic model.

PubMed

Vesely, Stepan; Klöckner, Christian A; Dohnal, Mirko

2016-03-01

In this paper we demonstrate that fuzzy logic can provide a better tool for predicting recycling behaviour than the customarily used linear regression. To show this, we take a set of empirical data on recycling behaviour (N=664), which we randomly divide into two halves. The first half is used to estimate a linear regression model of recycling behaviour, and to develop a fuzzy logic model of recycling behaviour. As the first comparison, the fit of both models to the data included in estimation of the models (N=332) is evaluated. As the second comparison, predictive accuracy of both models for "new" cases (hold-out data not included in building the models, N=332) is assessed. In both cases, the fuzzy logic model significantly outperforms the regression model in terms of fit. To conclude, when accurate predictions of recycling and possibly other environmental behaviours are needed, fuzzy logic modelling seems to be a promising technique. Copyright © 2015 Elsevier Ltd. All rights reserved.

Logistic random effects regression models: a comparison of statistical packages for binary and ordinal outcomes

PubMed Central

2011-01-01

Background Logistic random effects models are a popular tool to analyze multilevel also called hierarchical data with a binary or ordinal outcome. Here, we aim to compare different statistical software implementations of these models. Methods We used individual patient data from 8509 patients in 231 centers with moderate and severe Traumatic Brain Injury (TBI) enrolled in eight Randomized Controlled Trials (RCTs) and three observational studies. We fitted logistic random effects regression models with the 5-point Glasgow Outcome Scale (GOS) as outcome, both dichotomized as well as ordinal, with center and/or trial as random effects, and as covariates age, motor score, pupil reactivity or trial. We then compared the implementations of frequentist and Bayesian methods to estimate the fixed and random effects. Frequentist approaches included R (lme4), Stata (GLLAMM), SAS (GLIMMIX and NLMIXED), MLwiN ([R]IGLS) and MIXOR, Bayesian approaches included WinBUGS, MLwiN (MCMC), R package MCMCglmm and SAS experimental procedure MCMC. Three data sets (the full data set and two sub-datasets) were analysed using basically two logistic random effects models with either one random effect for the center or two random effects for center and trial. For the ordinal outcome in the full data set also a proportional odds model with a random center effect was fitted. Results The packages gave similar parameter estimates for both the fixed and random effects and for the binary (and ordinal) models for the main study and when based on a relatively large number of level-1 (patient level) data compared to the number of level-2 (hospital level) data. However, when based on relatively sparse data set, i.e. when the numbers of level-1 and level-2 data units were about the same, the frequentist and Bayesian approaches showed somewhat different results. The software implementations differ considerably in flexibility, computation time, and usability. There are also differences in the availability of additional tools for model evaluation, such as diagnostic plots. The experimental SAS (version 9.2) procedure MCMC appeared to be inefficient. Conclusions On relatively large data sets, the different software implementations of logistic random effects regression models produced similar results. Thus, for a large data set there seems to be no explicit preference (of course if there is no preference from a philosophical point of view) for either a frequentist or Bayesian approach (if based on vague priors). The choice for a particular implementation may largely depend on the desired flexibility, and the usability of the package. For small data sets the random effects variances are difficult to estimate. In the frequentist approaches the MLE of this variance was often estimated zero with a standard error that is either zero or could not be determined, while for Bayesian methods the estimates could depend on the chosen "non-informative" prior of the variance parameter. The starting value for the variance parameter may be also critical for the convergence of the Markov chain. PMID:21605357

Spatial prediction of landslides using a hybrid machine learning approach based on Random Subspace and Classification and Regression Trees

NASA Astrophysics Data System (ADS)

Pham, Binh Thai; Prakash, Indra; Tien Bui, Dieu

2018-02-01

A hybrid machine learning approach of Random Subspace (RSS) and Classification And Regression Trees (CART) is proposed to develop a model named RSSCART for spatial prediction of landslides. This model is a combination of the RSS method which is known as an efficient ensemble technique and the CART which is a state of the art classifier. The Luc Yen district of Yen Bai province, a prominent landslide prone area of Viet Nam, was selected for the model development. Performance of the RSSCART model was evaluated through the Receiver Operating Characteristic (ROC) curve, statistical analysis methods, and the Chi Square test. Results were compared with other benchmark landslide models namely Support Vector Machines (SVM), single CART, Naïve Bayes Trees (NBT), and Logistic Regression (LR). In the development of model, ten important landslide affecting factors related with geomorphology, geology and geo-environment were considered namely slope angles, elevation, slope aspect, curvature, lithology, distance to faults, distance to rivers, distance to roads, and rainfall. Performance of the RSSCART model (AUC = 0.841) is the best compared with other popular landslide models namely SVM (0.835), single CART (0.822), NBT (0.821), and LR (0.723). These results indicate that performance of the RSSCART is a promising method for spatial landslide prediction.

Genetic Parameter Estimates for Metabolizing Two Common Pharmaceuticals in Swine.

PubMed

Howard, Jeremy T; Ashwell, Melissa S; Baynes, Ronald E; Brooks, James D; Yeatts, James L; Maltecca, Christian

2018-01-01

In livestock, the regulation of drugs used to treat livestock has received increased attention and it is currently unknown how much of the phenotypic variation in drug metabolism is due to the genetics of an animal. Therefore, the objective of the study was to determine the amount of phenotypic variation in fenbendazole and flunixin meglumine drug metabolism due to genetics. The population consisted of crossbred female and castrated male nursery pigs ( n = 198) that were sired by boars represented by four breeds. The animals were spread across nine batches. Drugs were administered intravenously and blood collected a minimum of 10 times over a 48 h period. Genetic parameters for the parent drug and metabolite concentration within each drug were estimated based on pharmacokinetics (PK) parameters or concentrations across time utilizing a random regression model. The PK parameters were estimated using a non-compartmental analysis. The PK model included fixed effects of sex and breed of sire along with random sire and batch effects. The random regression model utilized Legendre polynomials and included a fixed population concentration curve, sex, and breed of sire effects along with a random sire deviation from the population curve and batch effect. The sire effect included the intercept for all models except for the fenbendazole metabolite (i.e., intercept and slope). The mean heritability across PK parameters for the fenbendazole and flunixin meglumine parent drug (metabolite) was 0.15 (0.18) and 0.31 (0.40), respectively. For the parent drug (metabolite), the mean heritability across time was 0.27 (0.60) and 0.14 (0.44) for fenbendazole and flunixin meglumine, respectively. The errors surrounding the heritability estimates for the random regression model were smaller compared to estimates obtained from PK parameters. Across both the PK and plasma drug concentration across model, a moderate heritability was estimated. The model that utilized the plasma drug concentration across time resulted in estimates with a smaller standard error compared to models that utilized PK parameters. The current study found a low to moderate proportion of the phenotypic variation in metabolizing fenbendazole and flunixin meglumine that was explained by genetics in the current study.

Genetic Parameter Estimates for Metabolizing Two Common Pharmaceuticals in Swine

PubMed Central

Howard, Jeremy T.; Ashwell, Melissa S.; Baynes, Ronald E.; Brooks, James D.; Yeatts, James L.; Maltecca, Christian

2018-01-01

In livestock, the regulation of drugs used to treat livestock has received increased attention and it is currently unknown how much of the phenotypic variation in drug metabolism is due to the genetics of an animal. Therefore, the objective of the study was to determine the amount of phenotypic variation in fenbendazole and flunixin meglumine drug metabolism due to genetics. The population consisted of crossbred female and castrated male nursery pigs (n = 198) that were sired by boars represented by four breeds. The animals were spread across nine batches. Drugs were administered intravenously and blood collected a minimum of 10 times over a 48 h period. Genetic parameters for the parent drug and metabolite concentration within each drug were estimated based on pharmacokinetics (PK) parameters or concentrations across time utilizing a random regression model. The PK parameters were estimated using a non-compartmental analysis. The PK model included fixed effects of sex and breed of sire along with random sire and batch effects. The random regression model utilized Legendre polynomials and included a fixed population concentration curve, sex, and breed of sire effects along with a random sire deviation from the population curve and batch effect. The sire effect included the intercept for all models except for the fenbendazole metabolite (i.e., intercept and slope). The mean heritability across PK parameters for the fenbendazole and flunixin meglumine parent drug (metabolite) was 0.15 (0.18) and 0.31 (0.40), respectively. For the parent drug (metabolite), the mean heritability across time was 0.27 (0.60) and 0.14 (0.44) for fenbendazole and flunixin meglumine, respectively. The errors surrounding the heritability estimates for the random regression model were smaller compared to estimates obtained from PK parameters. Across both the PK and plasma drug concentration across model, a moderate heritability was estimated. The model that utilized the plasma drug concentration across time resulted in estimates with a smaller standard error compared to models that utilized PK parameters. The current study found a low to moderate proportion of the phenotypic variation in metabolizing fenbendazole and flunixin meglumine that was explained by genetics in the current study. PMID:29487615

Incorporation of prior information on parameters into nonlinear regression groundwater flow models: 1. Theory

USGS Publications Warehouse

Cooley, Richard L.

1982-01-01

Prior information on the parameters of a groundwater flow model can be used to improve parameter estimates obtained from nonlinear regression solution of a modeling problem. Two scales of prior information can be available: (1) prior information having known reliability (that is, bias and random error structure) and (2) prior information consisting of best available estimates of unknown reliability. A regression method that incorporates the second scale of prior information assumes the prior information to be fixed for any particular analysis to produce improved, although biased, parameter estimates. Approximate optimization of two auxiliary parameters of the formulation is used to help minimize the bias, which is almost always much smaller than that resulting from standard ridge regression. It is shown that if both scales of prior information are available, then a combined regression analysis may be made.

Using regression methods to estimate stream phosphorus loads at the Illinois River, Arkansas

USGS Publications Warehouse

Haggard, B.E.; Soerens, T.S.; Green, W.R.; Richards, R.P.

2003-01-01

The development of total maximum daily loads (TMDLs) requires evaluating existing constituent loads in streams. Accurate estimates of constituent loads are needed to calibrate watershed and reservoir models for TMDL development. The best approach to estimate constituent loads is high frequency sampling, particularly during storm events, and mass integration of constituents passing a point in a stream. Most often, resources are limited and discrete water quality samples are collected on fixed intervals and sometimes supplemented with directed sampling during storm events. When resources are limited, mass integration is not an accurate means to determine constituent loads and other load estimation techniques such as regression models are used. The objective of this work was to determine a minimum number of water-quality samples needed to provide constituent concentration data adequate to estimate constituent loads at a large stream. Twenty sets of water quality samples with and without supplemental storm samples were randomly selected at various fixed intervals from a database at the Illinois River, northwest Arkansas. The random sets were used to estimate total phosphorus (TP) loads using regression models. The regression-based annual TP loads were compared to the integrated annual TP load estimated using all the data. At a minimum, monthly sampling plus supplemental storm samples (six samples per year) was needed to produce a root mean square error of less than 15%. Water quality samples should be collected at least semi-monthly (every 15 days) in studies less than two years if seasonal time factors are to be used in the regression models. Annual TP loads estimated from independently collected discrete water quality samples further demonstrated the utility of using regression models to estimate annual TP loads in this stream system.

Identification of immune correlates of protection in Shigella infection by application of machine learning.

PubMed

Arevalillo, Jorge M; Sztein, Marcelo B; Kotloff, Karen L; Levine, Myron M; Simon, Jakub K

2017-10-01

Immunologic correlates of protection are important in vaccine development because they give insight into mechanisms of protection, assist in the identification of promising vaccine candidates, and serve as endpoints in bridging clinical vaccine studies. Our goal is the development of a methodology to identify immunologic correlates of protection using the Shigella challenge as a model. The proposed methodology utilizes the Random Forests (RF) machine learning algorithm as well as Classification and Regression Trees (CART) to detect immune markers that predict protection, identify interactions between variables, and define optimal cutoffs. Logistic regression modeling is applied to estimate the probability of protection and the confidence interval (CI) for such a probability is computed by bootstrapping the logistic regression models. The results demonstrate that the combination of Classification and Regression Trees and Random Forests complements the standard logistic regression and uncovers subtle immune interactions. Specific levels of immunoglobulin IgG antibody in blood on the day of challenge predicted protection in 75% (95% CI 67-86). Of those subjects that did not have blood IgG at or above a defined threshold, 100% were protected if they had IgA antibody secreting cells above a defined threshold. Comparison with the results obtained by applying only logistic regression modeling with standard Akaike Information Criterion for model selection shows the usefulness of the proposed method. Given the complexity of the immune system, the use of machine learning methods may enhance traditional statistical approaches. When applied together, they offer a novel way to quantify important immune correlates of protection that may help the development of vaccines. Copyright © 2017 Elsevier Inc. All rights reserved.

A Bayesian Nonparametric Causal Model for Regression Discontinuity Designs

ERIC Educational Resources Information Center

Karabatsos, George; Walker, Stephen G.

2013-01-01

The regression discontinuity (RD) design (Thistlewaite & Campbell, 1960; Cook, 2008) provides a framework to identify and estimate causal effects from a non-randomized design. Each subject of a RD design is assigned to the treatment (versus assignment to a non-treatment) whenever her/his observed value of the assignment variable equals or…

Application of linear regression analysis in accuracy assessment of rolling force calculations

NASA Astrophysics Data System (ADS)

Poliak, E. I.; Shim, M. K.; Kim, G. S.; Choo, W. Y.

1998-10-01

Efficient operation of the computational models employed in process control systems require periodical assessment of the accuracy of their predictions. Linear regression is proposed as a tool which allows separate systematic and random prediction errors from those related to measurements. A quantitative characteristic of the model predictive ability is introduced in addition to standard statistical tests for model adequacy. Rolling force calculations are considered as an example for the application. However, the outlined approach can be used to assess the performance of any computational model.

Minimizing effects of methodological decisions on interpretation and prediction in species distribution studies: An example with background selection

USGS Publications Warehouse

Jarnevich, Catherine S.; Talbert, Marian; Morisette, Jeffrey T.; Aldridge, Cameron L.; Brown, Cynthia; Kumar, Sunil; Manier, Daniel; Talbert, Colin; Holcombe, Tracy R.

2017-01-01

Evaluating the conditions where a species can persist is an important question in ecology both to understand tolerances of organisms and to predict distributions across landscapes. Presence data combined with background or pseudo-absence locations are commonly used with species distribution modeling to develop these relationships. However, there is not a standard method to generate background or pseudo-absence locations, and method choice affects model outcomes. We evaluated combinations of both model algorithms (simple and complex generalized linear models, multivariate adaptive regression splines, Maxent, boosted regression trees, and random forest) and background methods (random, minimum convex polygon, and continuous and binary kernel density estimator (KDE)) to assess the sensitivity of model outcomes to choices made. We evaluated six questions related to model results, including five beyond the common comparison of model accuracy assessment metrics (biological interpretability of response curves, cross-validation robustness, independent data accuracy and robustness, and prediction consistency). For our case study with cheatgrass in the western US, random forest was least sensitive to background choice and the binary KDE method was least sensitive to model algorithm choice. While this outcome may not hold for other locations or species, the methods we used can be implemented to help determine appropriate methodologies for particular research questions.

Encrypted data stream identification using randomness sparse representation and fuzzy Gaussian mixture model

NASA Astrophysics Data System (ADS)

Zhang, Hong; Hou, Rui; Yi, Lei; Meng, Juan; Pan, Zhisong; Zhou, Yuhuan

2016-07-01

The accurate identification of encrypted data stream helps to regulate illegal data, detect network attacks and protect users' information. In this paper, a novel encrypted data stream identification algorithm is introduced. The proposed method is based on randomness characteristics of encrypted data stream. We use a l1-norm regularized logistic regression to improve sparse representation of randomness features and Fuzzy Gaussian Mixture Model (FGMM) to improve identification accuracy. Experimental results demonstrate that the method can be adopted as an effective technique for encrypted data stream identification.

Statistical methods for efficient design of community surveys of response to noise: Random coefficients regression models

NASA Technical Reports Server (NTRS)

Tomberlin, T. J.

1985-01-01

Research studies of residents' responses to noise consist of interviews with samples of individuals who are drawn from a number of different compact study areas. The statistical techniques developed provide a basis for those sample design decisions. These techniques are suitable for a wide range of sample survey applications. A sample may consist of a random sample of residents selected from a sample of compact study areas, or in a more complex design, of a sample of residents selected from a sample of larger areas (e.g., cities). The techniques may be applied to estimates of the effects on annoyance of noise level, numbers of noise events, the time-of-day of the events, ambient noise levels, or other factors. Methods are provided for determining, in advance, how accurately these effects can be estimated for different sample sizes and study designs. Using a simple cost function, they also provide for optimum allocation of the sample across the stages of the design for estimating these effects. These techniques are developed via a regression model in which the regression coefficients are assumed to be random, with components of variance associated with the various stages of a multi-stage sample design.

The Bayesian group lasso for confounded spatial data

USGS Publications Warehouse

Hefley, Trevor J.; Hooten, Mevin B.; Hanks, Ephraim M.; Russell, Robin E.; Walsh, Daniel P.

2017-01-01

Generalized linear mixed models for spatial processes are widely used in applied statistics. In many applications of the spatial generalized linear mixed model (SGLMM), the goal is to obtain inference about regression coefficients while achieving optimal predictive ability. When implementing the SGLMM, multicollinearity among covariates and the spatial random effects can make computation challenging and influence inference. We present a Bayesian group lasso prior with a single tuning parameter that can be chosen to optimize predictive ability of the SGLMM and jointly regularize the regression coefficients and spatial random effect. We implement the group lasso SGLMM using efficient Markov chain Monte Carlo (MCMC) algorithms and demonstrate how multicollinearity among covariates and the spatial random effect can be monitored as a derived quantity. To test our method, we compared several parameterizations of the SGLMM using simulated data and two examples from plant ecology and disease ecology. In all examples, problematic levels multicollinearity occurred and influenced sampling efficiency and inference. We found that the group lasso prior resulted in roughly twice the effective sample size for MCMC samples of regression coefficients and can have higher and less variable predictive accuracy based on out-of-sample data when compared to the standard SGLMM.

Analysis of training sample selection strategies for regression-based quantitative landslide susceptibility mapping methods

NASA Astrophysics Data System (ADS)

Erener, Arzu; Sivas, A. Abdullah; Selcuk-Kestel, A. Sevtap; Düzgün, H. Sebnem

2017-07-01

All of the quantitative landslide susceptibility mapping (QLSM) methods requires two basic data types, namely, landslide inventory and factors that influence landslide occurrence (landslide influencing factors, LIF). Depending on type of landslides, nature of triggers and LIF, accuracy of the QLSM methods differs. Moreover, how to balance the number of 0 (nonoccurrence) and 1 (occurrence) in the training set obtained from the landslide inventory and how to select which one of the 1's and 0's to be included in QLSM models play critical role in the accuracy of the QLSM. Although performance of various QLSM methods is largely investigated in the literature, the challenge of training set construction is not adequately investigated for the QLSM methods. In order to tackle this challenge, in this study three different training set selection strategies along with the original data set is used for testing the performance of three different regression methods namely Logistic Regression (LR), Bayesian Logistic Regression (BLR) and Fuzzy Logistic Regression (FLR). The first sampling strategy is proportional random sampling (PRS), which takes into account a weighted selection of landslide occurrences in the sample set. The second method, namely non-selective nearby sampling (NNS), includes randomly selected sites and their surrounding neighboring points at certain preselected distances to include the impact of clustering. Selective nearby sampling (SNS) is the third method, which concentrates on the group of 1's and their surrounding neighborhood. A randomly selected group of landslide sites and their neighborhood are considered in the analyses similar to NNS parameters. It is found that LR-PRS, FLR-PRS and BLR-Whole Data set-ups, with order, yield the best fits among the other alternatives. The results indicate that in QLSM based on regression models, avoidance of spatial correlation in the data set is critical for the model's performance.

Predicting 30-day Hospital Readmission with Publicly Available Administrative Database. A Conditional Logistic Regression Modeling Approach.

PubMed

Zhu, K; Lou, Z; Zhou, J; Ballester, N; Kong, N; Parikh, P

2015-01-01

This article is part of the Focus Theme of Methods of Information in Medicine on "Big Data and Analytics in Healthcare". Hospital readmissions raise healthcare costs and cause significant distress to providers and patients. It is, therefore, of great interest to healthcare organizations to predict what patients are at risk to be readmitted to their hospitals. However, current logistic regression based risk prediction models have limited prediction power when applied to hospital administrative data. Meanwhile, although decision trees and random forests have been applied, they tend to be too complex to understand among the hospital practitioners. Explore the use of conditional logistic regression to increase the prediction accuracy. We analyzed an HCUP statewide inpatient discharge record dataset, which includes patient demographics, clinical and care utilization data from California. We extracted records of heart failure Medicare beneficiaries who had inpatient experience during an 11-month period. We corrected the data imbalance issue with under-sampling. In our study, we first applied standard logistic regression and decision tree to obtain influential variables and derive practically meaning decision rules. We then stratified the original data set accordingly and applied logistic regression on each data stratum. We further explored the effect of interacting variables in the logistic regression modeling. We conducted cross validation to assess the overall prediction performance of conditional logistic regression (CLR) and compared it with standard classification models. The developed CLR models outperformed several standard classification models (e.g., straightforward logistic regression, stepwise logistic regression, random forest, support vector machine). For example, the best CLR model improved the classification accuracy by nearly 20% over the straightforward logistic regression model. Furthermore, the developed CLR models tend to achieve better sensitivity of more than 10% over the standard classification models, which can be translated to correct labeling of additional 400 - 500 readmissions for heart failure patients in the state of California over a year. Lastly, several key predictor identified from the HCUP data include the disposition location from discharge, the number of chronic conditions, and the number of acute procedures. It would be beneficial to apply simple decision rules obtained from the decision tree in an ad-hoc manner to guide the cohort stratification. It could be potentially beneficial to explore the effect of pairwise interactions between influential predictors when building the logistic regression models for different data strata. Judicious use of the ad-hoc CLR models developed offers insights into future development of prediction models for hospital readmissions, which can lead to better intuition in identifying high-risk patients and developing effective post-discharge care strategies. Lastly, this paper is expected to raise the awareness of collecting data on additional markers and developing necessary database infrastructure for larger-scale exploratory studies on readmission risk prediction.

Epidemiological characteristics of reported sporadic and outbreak cases of E. coli O157 in people from Alberta, Canada (2000-2002): methodological challenges of comparing clustered to unclustered data.

PubMed

Pearl, D L; Louie, M; Chui, L; Doré, K; Grimsrud, K M; Martin, S W; Michel, P; Svenson, L W; McEwen, S A

2008-04-01

Using multivariable models, we compared whether there were significant differences between reported outbreak and sporadic cases in terms of their sex, age, and mode and site of disease transmission. We also determined the potential role of administrative, temporal, and spatial factors within these models. We compared a variety of approaches to account for clustering of cases in outbreaks including weighted logistic regression, random effects models, general estimating equations, robust variance estimates, and the random selection of one case from each outbreak. Age and mode of transmission were the only epidemiologically and statistically significant covariates in our final models using the above approaches. Weighing observations in a logistic regression model by the inverse of their outbreak size appeared to be a relatively robust and valid means for modelling these data. Some analytical techniques, designed to account for clustering, had difficulty converging or producing realistic measures of association.

Modelling subject-specific childhood growth using linear mixed-effect models with cubic regression splines.

PubMed

Grajeda, Laura M; Ivanescu, Andrada; Saito, Mayuko; Crainiceanu, Ciprian; Jaganath, Devan; Gilman, Robert H; Crabtree, Jean E; Kelleher, Dermott; Cabrera, Lilia; Cama, Vitaliano; Checkley, William

2016-01-01

Childhood growth is a cornerstone of pediatric research. Statistical models need to consider individual trajectories to adequately describe growth outcomes. Specifically, well-defined longitudinal models are essential to characterize both population and subject-specific growth. Linear mixed-effect models with cubic regression splines can account for the nonlinearity of growth curves and provide reasonable estimators of population and subject-specific growth, velocity and acceleration. We provide a stepwise approach that builds from simple to complex models, and account for the intrinsic complexity of the data. We start with standard cubic splines regression models and build up to a model that includes subject-specific random intercepts and slopes and residual autocorrelation. We then compared cubic regression splines vis-à-vis linear piecewise splines, and with varying number of knots and positions. Statistical code is provided to ensure reproducibility and improve dissemination of methods. Models are applied to longitudinal height measurements in a cohort of 215 Peruvian children followed from birth until their fourth year of life. Unexplained variability, as measured by the variance of the regression model, was reduced from 7.34 when using ordinary least squares to 0.81 (p < 0.001) when using a linear mixed-effect models with random slopes and a first order continuous autoregressive error term. There was substantial heterogeneity in both the intercept (p < 0.001) and slopes (p < 0.001) of the individual growth trajectories. We also identified important serial correlation within the structure of the data (ρ = 0.66; 95 % CI 0.64 to 0.68; p < 0.001), which we modeled with a first order continuous autoregressive error term as evidenced by the variogram of the residuals and by a lack of association among residuals. The final model provides a parametric linear regression equation for both estimation and prediction of population- and individual-level growth in height. We show that cubic regression splines are superior to linear regression splines for the case of a small number of knots in both estimation and prediction with the full linear mixed effect model (AIC 19,352 vs. 19,598, respectively). While the regression parameters are more complex to interpret in the former, we argue that inference for any problem depends more on the estimated curve or differences in curves rather than the coefficients. Moreover, use of cubic regression splines provides biological meaningful growth velocity and acceleration curves despite increased complexity in coefficient interpretation. Through this stepwise approach, we provide a set of tools to model longitudinal childhood data for non-statisticians using linear mixed-effect models.

Auto Regressive Moving Average (ARMA) Modeling Method for Gyro Random Noise Using a Robust Kalman Filter

PubMed Central

Huang, Lei

2015-01-01

To solve the problem in which the conventional ARMA modeling methods for gyro random noise require a large number of samples and converge slowly, an ARMA modeling method using a robust Kalman filtering is developed. The ARMA model parameters are employed as state arguments. Unknown time-varying estimators of observation noise are used to achieve the estimated mean and variance of the observation noise. Using the robust Kalman filtering, the ARMA model parameters are estimated accurately. The developed ARMA modeling method has the advantages of a rapid convergence and high accuracy. Thus, the required sample size is reduced. It can be applied to modeling applications for gyro random noise in which a fast and accurate ARMA modeling method is required. PMID:26437409

[Application of negative binomial regression and modified Poisson regression in the research of risk factors for injury frequency].

PubMed

Cao, Qingqing; Wu, Zhenqiang; Sun, Ying; Wang, Tiezhu; Han, Tengwei; Gu, Chaomei; Sun, Yehuan

2011-11-01

To Eexplore the application of negative binomial regression and modified Poisson regression analysis in analyzing the influential factors for injury frequency and the risk factors leading to the increase of injury frequency. 2917 primary and secondary school students were selected from Hefei by cluster random sampling method and surveyed by questionnaire. The data on the count event-based injuries used to fitted modified Poisson regression and negative binomial regression model. The risk factors incurring the increase of unintentional injury frequency for juvenile students was explored, so as to probe the efficiency of these two models in studying the influential factors for injury frequency. The Poisson model existed over-dispersion (P < 0.0001) based on testing by the Lagrangemultiplier. Therefore, the over-dispersion dispersed data using a modified Poisson regression and negative binomial regression model, was fitted better. respectively. Both showed that male gender, younger age, father working outside of the hometown, the level of the guardian being above junior high school and smoking might be the results of higher injury frequencies. On a tendency of clustered frequency data on injury event, both the modified Poisson regression analysis and negative binomial regression analysis can be used. However, based on our data, the modified Poisson regression fitted better and this model could give a more accurate interpretation of relevant factors affecting the frequency of injury.

Multivariate random regression analysis for body weight and main morphological traits in genetically improved farmed tilapia (Oreochromis niloticus).

PubMed

He, Jie; Zhao, Yunfeng; Zhao, Jingli; Gao, Jin; Han, Dandan; Xu, Pao; Yang, Runqing

2017-11-02

Because of their high economic importance, growth traits in fish are under continuous improvement. For growth traits that are recorded at multiple time-points in life, the use of univariate and multivariate animal models is limited because of the variable and irregular timing of these measures. Thus, the univariate random regression model (RRM) was introduced for the genetic analysis of dynamic growth traits in fish breeding. We used a multivariate random regression model (MRRM) to analyze genetic changes in growth traits recorded at multiple time-point of genetically-improved farmed tilapia. Legendre polynomials of different orders were applied to characterize the influences of fixed and random effects on growth trajectories. The final MRRM was determined by optimizing the univariate RRM for the analyzed traits separately via penalizing adaptively the likelihood statistical criterion, which is superior to both the Akaike information criterion and the Bayesian information criterion. In the selected MRRM, the additive genetic effects were modeled by Legendre polynomials of three orders for body weight (BWE) and body length (BL) and of two orders for body depth (BD). By using the covariance functions of the MRRM, estimated heritabilities were between 0.086 and 0.628 for BWE, 0.155 and 0.556 for BL, and 0.056 and 0.607 for BD. Only heritabilities for BD measured from 60 to 140 days of age were consistently higher than those estimated by the univariate RRM. All genetic correlations between growth time-points exceeded 0.5 for either single or pairwise time-points. Moreover, correlations between early and late growth time-points were lower. Thus, for phenotypes that are measured repeatedly in aquaculture, an MRRM can enhance the efficiency of the comprehensive selection for BWE and the main morphological traits.

Development of an anaerobic threshold (HRLT, HRVT) estimation equation using the heart rate threshold (HRT) during the treadmill incremental exercise test

PubMed Central

Ham, Joo-ho; Park, Hun-Young; Kim, Youn-ho; Bae, Sang-kon; Ko, Byung-hoon

2017-01-01

[Purpose] The purpose of this study was to develop a regression model to estimate the heart rate at the lactate threshold (HRLT) and the heart rate at the ventilatory threshold (HRVT) using the heart rate threshold (HRT), and to test the validity of the regression model. [Methods] We performed a graded exercise test with a treadmill in 220 normal individuals (men: 112, women: 108) aged 20–59 years. HRT, HRLT, and HRVT were measured in all subjects. A regression model was developed to estimate HRLT and HRVT using HRT with 70% of the data (men: 79, women: 76) through randomization (7:3), with the Bernoulli trial. The validity of the regression model developed with the remaining 30% of the data (men: 33, women: 32) was also examined. [Results] Based on the regression coefficient, we found that the independent variable HRT was a significant variable in all regression models. The adjusted R2 of the developed regression models averaged about 70%, and the standard error of estimation of the validity test results was 11 bpm, which is similar to that of the developed model. [Conclusion] These results suggest that HRT is a useful parameter for predicting HRLT and HRVT. PMID:29036765

Development of an anaerobic threshold (HRLT, HRVT) estimation equation using the heart rate threshold (HRT) during the treadmill incremental exercise test.

PubMed

Ham, Joo-Ho; Park, Hun-Young; Kim, Youn-Ho; Bae, Sang-Kon; Ko, Byung-Hoon; Nam, Sang-Seok

2017-09-30

The purpose of this study was to develop a regression model to estimate the heart rate at the lactate threshold (HRLT) and the heart rate at the ventilatory threshold (HRVT) using the heart rate threshold (HRT), and to test the validity of the regression model. We performed a graded exercise test with a treadmill in 220 normal individuals (men: 112, women: 108) aged 20-59 years. HRT, HRLT, and HRVT were measured in all subjects. A regression model was developed to estimate HRLT and HRVT using HRT with 70% of the data (men: 79, women: 76) through randomization (7:3), with the Bernoulli trial. The validity of the regression model developed with the remaining 30% of the data (men: 33, women: 32) was also examined. Based on the regression coefficient, we found that the independent variable HRT was a significant variable in all regression models. The adjusted R2 of the developed regression models averaged about 70%, and the standard error of estimation of the validity test results was 11 bpm, which is similar to that of the developed model. These results suggest that HRT is a useful parameter for predicting HRLT and HRVT. ©2017 The Korean Society for Exercise Nutrition

Measures of Residual Risk with Connections to Regression, Risk Tracking, Surrogate Models, and Ambiguity

DTIC Science & Technology

2015-01-07

vector that helps to manage , predict, and mitigate the risk in the original variable. Residual risk can be exemplified as a quantification of the improved... the random variable of interest is viewed in concert with a related random vector that helps to manage , predict, and mitigate the risk in the original...measures of risk. They view a random variable of interest in concert with an auxiliary random vector that helps to manage , predict and mitigate the risk

Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses.

PubMed

Faul, Franz; Erdfelder, Edgar; Buchner, Axel; Lang, Albert-Georg

2009-11-01

G*Power is a free power analysis program for a variety of statistical tests. We present extensions and improvements of the version introduced by Faul, Erdfelder, Lang, and Buchner (2007) in the domain of correlation and regression analyses. In the new version, we have added procedures to analyze the power of tests based on (1) single-sample tetrachoric correlations, (2) comparisons of dependent correlations, (3) bivariate linear regression, (4) multiple linear regression based on the random predictor model, (5) logistic regression, and (6) Poisson regression. We describe these new features and provide a brief introduction to their scope and handling.

Panel regressions to estimate low-flow response to rainfall variability in ungaged basins

USGS Publications Warehouse

Bassiouni, Maoya; Vogel, Richard M.; Archfield, Stacey A.

2016-01-01

Multicollinearity and omitted-variable bias are major limitations to developing multiple linear regression models to estimate streamflow characteristics in ungaged areas and varying rainfall conditions. Panel regression is used to overcome limitations of traditional regression methods, and obtain reliable model coefficients, in particular to understand the elasticity of streamflow to rainfall. Using annual rainfall and selected basin characteristics at 86 gaged streams in the Hawaiian Islands, regional regression models for three stream classes were developed to estimate the annual low-flow duration discharges. Three panel-regression structures (random effects, fixed effects, and pooled) were compared to traditional regression methods, in which space is substituted for time. Results indicated that panel regression generally was able to reproduce the temporal behavior of streamflow and reduce the standard errors of model coefficients compared to traditional regression, even for models in which the unobserved heterogeneity between streams is significant and the variance inflation factor for rainfall is much greater than 10. This is because both spatial and temporal variability were better characterized in panel regression. In a case study, regional rainfall elasticities estimated from panel regressions were applied to ungaged basins on Maui, using available rainfall projections to estimate plausible changes in surface-water availability and usable stream habitat for native species. The presented panel-regression framework is shown to offer benefits over existing traditional hydrologic regression methods for developing robust regional relations to investigate streamflow response in a changing climate.

Panel regressions to estimate low-flow response to rainfall variability in ungaged basins

NASA Astrophysics Data System (ADS)

Bassiouni, Maoya; Vogel, Richard M.; Archfield, Stacey A.

2016-12-01

Multicollinearity and omitted-variable bias are major limitations to developing multiple linear regression models to estimate streamflow characteristics in ungaged areas and varying rainfall conditions. Panel regression is used to overcome limitations of traditional regression methods, and obtain reliable model coefficients, in particular to understand the elasticity of streamflow to rainfall. Using annual rainfall and selected basin characteristics at 86 gaged streams in the Hawaiian Islands, regional regression models for three stream classes were developed to estimate the annual low-flow duration discharges. Three panel-regression structures (random effects, fixed effects, and pooled) were compared to traditional regression methods, in which space is substituted for time. Results indicated that panel regression generally was able to reproduce the temporal behavior of streamflow and reduce the standard errors of model coefficients compared to traditional regression, even for models in which the unobserved heterogeneity between streams is significant and the variance inflation factor for rainfall is much greater than 10. This is because both spatial and temporal variability were better characterized in panel regression. In a case study, regional rainfall elasticities estimated from panel regressions were applied to ungaged basins on Maui, using available rainfall projections to estimate plausible changes in surface-water availability and usable stream habitat for native species. The presented panel-regression framework is shown to offer benefits over existing traditional hydrologic regression methods for developing robust regional relations to investigate streamflow response in a changing climate.

Use and interpretation of logistic regression in habitat-selection studies

USGS Publications Warehouse

Keating, Kim A.; Cherry, Steve

2004-01-01

 Logistic regression is an important tool for wildlife habitat-selection studies, but the method frequently has been misapplied due to an inadequate understanding of the logistic model, its interpretation, and the influence of sampling design. To promote better use of this method, we review its application and interpretation under 3 sampling designs: random, case-control, and use-availability. Logistic regression is appropriate for habitat use-nonuse studies employing random sampling and can be used to directly model the conditional probability of use in such cases. Logistic regression also is appropriate for studies employing case-control sampling designs, but careful attention is required to interpret results correctly. Unless bias can be estimated or probability of use is small for all habitats, results of case-control studies should be interpreted as odds ratios, rather than probability of use or relative probability of use. When data are gathered under a use-availability design, logistic regression can be used to estimate approximate odds ratios if probability of use is small, at least on average. More generally, however, logistic regression is inappropriate for modeling habitat selection in use-availability studies. In particular, using logistic regression to fit the exponential model of Manly et al. (2002:100) does not guarantee maximum-likelihood estimates, valid probabilities, or valid likelihoods. We show that the resource selection function (RSF) commonly used for the exponential model is proportional to a logistic discriminant function. Thus, it may be used to rank habitats with respect to probability of use and to identify important habitat characteristics or their surrogates, but it is not guaranteed to be proportional to probability of use. Other problems associated with the exponential model also are discussed. We describe an alternative model based on Lancaster and Imbens (1996) that offers a method for estimating conditional probability of use in use-availability studies. Although promising, this model fails to converge to a unique solution in some important situations. Further work is needed to obtain a robust method that is broadly applicable to use-availability studies.

Representation of limb kinematics in Purkinje cell simple spike discharge is conserved across multiple tasks.

PubMed

Hewitt, Angela L; Popa, Laurentiu S; Pasalar, Siavash; Hendrix, Claudia M; Ebner, Timothy J

2011-11-01

Encoding of movement kinematics in Purkinje cell simple spike discharge has important implications for hypotheses of cerebellar cortical function. Several outstanding questions remain regarding representation of these kinematic signals. It is uncertain whether kinematic encoding occurs in unpredictable, feedback-dependent tasks or kinematic signals are conserved across tasks. Additionally, there is a need to understand the signals encoded in the instantaneous discharge of single cells without averaging across trials or time. To address these questions, this study recorded Purkinje cell firing in monkeys trained to perform a manual random tracking task in addition to circular tracking and center-out reach. Random tracking provides for extensive coverage of kinematic workspaces. Direction and speed errors are significantly greater during random than circular tracking. Cross-correlation analyses comparing hand and target velocity profiles show that hand velocity lags target velocity during random tracking. Correlations between simple spike firing from 120 Purkinje cells and hand position, velocity, and speed were evaluated with linear regression models including a time constant, τ, as a measure of the firing lead/lag relative to the kinematic parameters. Across the population, velocity accounts for the majority of simple spike firing variability (63 ± 30% of R(adj)(2)), followed by position (28 ± 24% of R(adj)(2)) and speed (11 ± 19% of R(adj)(2)). Simple spike firing often leads hand kinematics. Comparison of regression models based on averaged vs. nonaveraged firing and kinematics reveals lower R(adj)(2) values for nonaveraged data; however, regression coefficients and τ values are highly similar. Finally, for most cells, model coefficients generated from random tracking accurately estimate simple spike firing in either circular tracking or center-out reach. These findings imply that the cerebellum controls movement kinematics, consistent with a forward internal model that predicts upcoming limb kinematics.

Strengthening the Regression Discontinuity Design Using Additional Design Elements: A Within-Study Comparison

ERIC Educational Resources Information Center

Wing, Coady; Cook, Thomas D.

2013-01-01

The sharp regression discontinuity design (RDD) has three key weaknesses compared to the randomized clinical trial (RCT). It has lower statistical power, it is more dependent on statistical modeling assumptions, and its treatment effect estimates are limited to the narrow subpopulation of cases immediately around the cutoff, which is rarely of…

Large unbalanced credit scoring using Lasso-logistic regression ensemble.

PubMed

Wang, Hong; Xu, Qingsong; Zhou, Lifeng

2015-01-01

Recently, various ensemble learning methods with different base classifiers have been proposed for credit scoring problems. However, for various reasons, there has been little research using logistic regression as the base classifier. In this paper, given large unbalanced data, we consider the plausibility of ensemble learning using regularized logistic regression as the base classifier to deal with credit scoring problems. In this research, the data is first balanced and diversified by clustering and bagging algorithms. Then we apply a Lasso-logistic regression learning ensemble to evaluate the credit risks. We show that the proposed algorithm outperforms popular credit scoring models such as decision tree, Lasso-logistic regression and random forests in terms of AUC and F-measure. We also provide two importance measures for the proposed model to identify important variables in the data.

Genetic parameters for body condition score, body weight, milk yield, and fertility estimated using random regression models.

PubMed

Berry, D P; Buckley, F; Dillon, P; Evans, R D; Rath, M; Veerkamp, R F

2003-11-01

Genetic (co)variances between body condition score (BCS), body weight (BW), milk yield, and fertility were estimated using a random regression animal model extended to multivariate analysis. The data analyzed included 81,313 BCS observations, 91,937 BW observations, and 100,458 milk test-day yields from 8725 multiparous Holstein-Friesian cows. A cubic random regression was sufficient to model the changing genetic variances for BCS, BW, and milk across different days in milk. The genetic correlations between BCS and fertility changed little over the lactation; genetic correlations between BCS and interval to first service and between BCS and pregnancy rate to first service varied from -0.47 to -0.31, and from 0.15 to 0.38, respectively. This suggests that maximum genetic gain in fertility from indirect selection on BCS should be based on measurements taken in midlactation when the genetic variance for BCS is largest. Selection for increased BW resulted in shorter intervals to first service, but more services and poorer pregnancy rates; genetic correlations between BW and pregnancy rate to first service varied from -0.52 to -0.45. Genetic selection for higher lactation milk yield alone through selection on increased milk yield in early lactation is likely to have a more deleterious effect on genetic merit for fertility than selection on higher milk yield in late lactation.

Random regression models using different functions to model test-day milk yield of Brazilian Holstein cows.

PubMed

Bignardi, A B; El Faro, L; Torres Júnior, R A A; Cardoso, V L; Machado, P F; Albuquerque, L G

2011-10-31

We analyzed 152,145 test-day records from 7317 first lactations of Holstein cows recorded from 1995 to 2003. Our objective was to model variations in test-day milk yield during the first lactation of Holstein cows by random regression model (RRM), using various functions in order to obtain adequate and parsimonious models for the estimation of genetic parameters. Test-day milk yields were grouped into weekly classes of days in milk, ranging from 1 to 44 weeks. The contemporary groups were defined as herd-test-day. The analyses were performed using a single-trait RRM, including the direct additive, permanent environmental and residual random effects. In addition, contemporary group and linear and quadratic effects of the age of cow at calving were included as fixed effects. The mean trend of milk yield was modeled with a fourth-order orthogonal Legendre polynomial. The additive genetic and permanent environmental covariance functions were estimated by random regression on two parametric functions, Ali and Schaeffer and Wilmink, and on B-spline functions of days in milk. The covariance components and the genetic parameters were estimated by the restricted maximum likelihood method. Results from RRM parametric and B-spline functions were compared to RRM on Legendre polynomials and with a multi-trait analysis, using the same data set. Heritability estimates presented similar trends during mid-lactation (13 to 31 weeks) and between week 37 and the end of lactation, for all RRM. Heritabilities obtained by multi-trait analysis were of a lower magnitude than those estimated by RRM. The RRMs with a higher number of parameters were more useful to describe the genetic variation of test-day milk yield throughout the lactation. RRM using B-spline and Legendre polynomials as base functions appears to be the most adequate to describe the covariance structure of the data.

The arcsine is asinine: the analysis of proportions in ecology.

PubMed

Warton, David I; Hui, Francis K C

2011-01-01

The arcsine square root transformation has long been standard procedure when analyzing proportional data in ecology, with applications in data sets containing binomial and non-binomial response variables. Here, we argue that the arcsine transform should not be used in either circumstance. For binomial data, logistic regression has greater interpretability and higher power than analyses of transformed data. However, it is important to check the data for additional unexplained variation, i.e., overdispersion, and to account for it via the inclusion of random effects in the model if found. For non-binomial data, the arcsine transform is undesirable on the grounds of interpretability, and because it can produce nonsensical predictions. The logit transformation is proposed as an alternative approach to address these issues. Examples are presented in both cases to illustrate these advantages, comparing various methods of analyzing proportions including untransformed, arcsine- and logit-transformed linear models and logistic regression (with or without random effects). Simulations demonstrate that logistic regression usually provides a gain in power over other methods.

Analysis of Palm Oil Production, Export, and Government Consumption to Gross Domestic Product of Five Districts in West Kalimantan by Panel Regression

NASA Astrophysics Data System (ADS)

Sulistianingsih, E.; Kiftiah, M.; Rosadi, D.; Wahyuni, H.

2017-04-01

Gross Domestic Product (GDP) is an indicator of economic growth in a region. GDP is a panel data, which consists of cross-section and time series data. Meanwhile, panel regression is a tool which can be utilised to analyse panel data. There are three models in panel regression, namely Common Effect Model (CEM), Fixed Effect Model (FEM) and Random Effect Model (REM). The models will be chosen based on results of Chow Test, Hausman Test and Lagrange Multiplier Test. This research analyses palm oil about production, export, and government consumption to five district GDP are in West Kalimantan, namely Sanggau, Sintang, Sambas, Ketapang and Bengkayang by panel regression. Based on the results of analyses, it concluded that REM, which adjusted-determination-coefficient is 0,823, is the best model in this case. Also, according to the result, only Export and Government Consumption that influence GDP of the districts.

Random effects coefficient of determination for mixed and meta-analysis models

PubMed Central

Demidenko, Eugene; Sargent, James; Onega, Tracy

2011-01-01

The key feature of a mixed model is the presence of random effects. We have developed a coefficient, called the random effects coefficient of determination, Rr2, that estimates the proportion of the conditional variance of the dependent variable explained by random effects. This coefficient takes values from 0 to 1 and indicates how strong the random effects are. The difference from the earlier suggested fixed effects coefficient of determination is emphasized. If Rr2 is close to 0, there is weak support for random effects in the model because the reduction of the variance of the dependent variable due to random effects is small; consequently, random effects may be ignored and the model simplifies to standard linear regression. The value of Rr2 apart from 0 indicates the evidence of the variance reduction in support of the mixed model. If random effects coefficient of determination is close to 1 the variance of random effects is very large and random effects turn into free fixed effects—the model can be estimated using the dummy variable approach. We derive explicit formulas for Rr2 in three special cases: the random intercept model, the growth curve model, and meta-analysis model. Theoretical results are illustrated with three mixed model examples: (1) travel time to the nearest cancer center for women with breast cancer in the U.S., (2) cumulative time watching alcohol related scenes in movies among young U.S. teens, as a risk factor for early drinking onset, and (3) the classic example of the meta-analysis model for combination of 13 studies on tuberculosis vaccine. PMID:23750070

A matrix-based method of moments for fitting the multivariate random effects model for meta-analysis and meta-regression

PubMed Central

Jackson, Dan; White, Ian R; Riley, Richard D

2013-01-01

Multivariate meta-analysis is becoming more commonly used. Methods for fitting the multivariate random effects model include maximum likelihood, restricted maximum likelihood, Bayesian estimation and multivariate generalisations of the standard univariate method of moments. Here, we provide a new multivariate method of moments for estimating the between-study covariance matrix with the properties that (1) it allows for either complete or incomplete outcomes and (2) it allows for covariates through meta-regression. Further, for complete data, it is invariant to linear transformations. Our method reduces to the usual univariate method of moments, proposed by DerSimonian and Laird, in a single dimension. We illustrate our method and compare it with some of the alternatives using a simulation study and a real example. PMID:23401213

An evaluation of supervised classifiers for indirectly detecting salt-affected areas at irrigation scheme level

NASA Astrophysics Data System (ADS)

Muller, Sybrand Jacobus; van Niekerk, Adriaan

2016-07-01

Soil salinity often leads to reduced crop yield and quality and can render soils barren. Irrigated areas are particularly at risk due to intensive cultivation and secondary salinization caused by waterlogging. Regular monitoring of salt accumulation in irrigation schemes is needed to keep its negative effects under control. The dynamic spatial and temporal characteristics of remote sensing can provide a cost-effective solution for monitoring salt accumulation at irrigation scheme level. This study evaluated a range of pan-fused SPOT-5 derived features (spectral bands, vegetation indices, image textures and image transformations) for classifying salt-affected areas in two distinctly different irrigation schemes in South Africa, namely Vaalharts and Breede River. The relationship between the input features and electro conductivity measurements were investigated using regression modelling (stepwise linear regression, partial least squares regression, curve fit regression modelling) and supervised classification (maximum likelihood, nearest neighbour, decision tree analysis, support vector machine and random forests). Classification and regression trees and random forest were used to select the most important features for differentiating salt-affected and unaffected areas. The results showed that the regression analyses produced weak models (<0.4 R squared). Better results were achieved using the supervised classifiers, but the algorithms tend to over-estimate salt-affected areas. A key finding was that none of the feature sets or classification algorithms stood out as being superior for monitoring salt accumulation at irrigation scheme level. This was attributed to the large variations in the spectral responses of different crops types at different growing stages, coupled with their individual tolerances to saline conditions.

Random regression analyses using B-splines to model growth of Australian Angus cattle

PubMed Central

Meyer, Karin

2005-01-01

Regression on the basis function of B-splines has been advocated as an alternative to orthogonal polynomials in random regression analyses. Basic theory of splines in mixed model analyses is reviewed, and estimates from analyses of weights of Australian Angus cattle from birth to 820 days of age are presented. Data comprised 84 533 records on 20 731 animals in 43 herds, with a high proportion of animals with 4 or more weights recorded. Changes in weights with age were modelled through B-splines of age at recording. A total of thirteen analyses, considering different combinations of linear, quadratic and cubic B-splines and up to six knots, were carried out. Results showed good agreement for all ages with many records, but fluctuated where data were sparse. On the whole, analyses using B-splines appeared more robust against "end-of-range" problems and yielded more consistent and accurate estimates of the first eigenfunctions than previous, polynomial analyses. A model fitting quadratic B-splines, with knots at 0, 200, 400, 600 and 821 days and a total of 91 covariance components, appeared to be a good compromise between detailedness of the model, number of parameters to be estimated, plausibility of results, and fit, measured as residual mean square error. PMID:16093011

Computer Simulation of Human Service Program Evaluations.

ERIC Educational Resources Information Center

Trochim, William M. K.; Davis, James E.

1985-01-01

Describes uses of computer simulations for the context of human service program evaluation. Presents simple mathematical models for most commonly used human service outcome evaluation designs (pretest-posttest randomized experiment, pretest-posttest nonequivalent groups design, and regression-discontinuity design). Translates models into single…

Reader reaction to "a robust method for estimating optimal treatment regimes" by Zhang et al. (2012).

PubMed

Taylor, Jeremy M G; Cheng, Wenting; Foster, Jared C

2015-03-01

A recent article (Zhang et al., 2012, Biometrics 168, 1010-1018) compares regression based and inverse probability based methods of estimating an optimal treatment regime and shows for a small number of covariates that inverse probability weighted methods are more robust to model misspecification than regression methods. We demonstrate that using models that fit the data better reduces the concern about non-robustness for the regression methods. We extend the simulation study of Zhang et al. (2012, Biometrics 168, 1010-1018), also considering the situation of a larger number of covariates, and show that incorporating random forests into both regression and inverse probability weighted based methods improves their properties. © 2014, The International Biometric Society.

[Modeling of an influence of indicators of social stress on demographic processes in regions of the Russian Federation].

PubMed

Burkin, M M; Molchanova, E V

To assess an impact of indicators of social stress on demographic processes in regions of the Russian Federation using statistical methods. The data of Rosstat «Regions of Russia» and «Health care in Russia» were used as information base. Indicators of about 80 subjects of the Russian Federation (without autonomous areas) for the ten-year period (2005-2014) have been created in the form of the database consisting of the following blocks: medico-demographic situation, level of economic development of the territory and wellbeing of the population, development of social infrastructure, ecological and climatic conditions, scientific researches and innovations. In total, there were about 70 indicators. Panel data for 80 regions of Russia in 10 years, which combine both indicators of spatial type (cross-section data), and information on temporary ranks (time-series data), were used. Various models of regression according to the panel data have been realized: the integrated model of regression (pooled model), regression model with the fixed effects (fixed effect model), regression model with random effects (random effect model). Main demographic indicators (life expectancy, birth rate, mortality from the external reasons) are to a great extent connected with socio-economic factors. Social tension (social stress) caused by transition to market economy plays an important role. The integral assessment of the impact of the average per capita monetary income, incidence of alcoholism and alcoholic psychoses, criminality, sales volume of alcoholic beverages per capita and marriage relations on demographic indicators is presented. Results of modeling allow to define the priority directions in the field of development of mental health and psychotherapeutic services in the regions of the Russian Federation.

Randomization at the level of primary care practice: use of pre-intervention data and random effects models.

PubMed

Nixon, R M; Duffy, S W; Fender, G R; Day, N E; Prevost, T C

2001-06-30

The Anglia menorrhagia education study tests the effectiveness of an education package for the treatment of menorrhagia given to doctors at a primary care level. General practices were randomized to receive or not receive the package. It is hoped that this intervention will reduce the proportion of women suffering from menorrhagia that are referred to hospital. Data are available on the treatment and referral of women in the practices in the education and control groups, both pre- and post-intervention. We define and demonstrate a random effects logistic regression model that includes pre-intervention data for calculating the effectiveness of the intervention. Copyright 2001 John Wiley & Sons, Ltd.

Regression Discontinuity Designs in Epidemiology

PubMed Central

Moscoe, Ellen; Mutevedzi, Portia; Newell, Marie-Louise; Bärnighausen, Till

2014-01-01

When patients receive an intervention based on whether they score below or above some threshold value on a continuously measured random variable, the intervention will be randomly assigned for patients close to the threshold. The regression discontinuity design exploits this fact to estimate causal treatment effects. In spite of its recent proliferation in economics, the regression discontinuity design has not been widely adopted in epidemiology. We describe regression discontinuity, its implementation, and the assumptions required for causal inference. We show that regression discontinuity is generalizable to the survival and nonlinear models that are mainstays of epidemiologic analysis. We then present an application of regression discontinuity to the much-debated epidemiologic question of when to start HIV patients on antiretroviral therapy. Using data from a large South African cohort (2007–2011), we estimate the causal effect of early versus deferred treatment eligibility on mortality. Patients whose first CD4 count was just below the 200 cells/μL CD4 count threshold had a 35% lower hazard of death (hazard ratio = 0.65 [95% confidence interval = 0.45–0.94]) than patients presenting with CD4 counts just above the threshold. We close by discussing the strengths and limitations of regression discontinuity designs for epidemiology. PMID:25061922

Decision tree modeling using R.

PubMed

Zhang, Zhongheng

2016-08-01

In machine learning field, decision tree learner is powerful and easy to interpret. It employs recursive binary partitioning algorithm that splits the sample in partitioning variable with the strongest association with the response variable. The process continues until some stopping criteria are met. In the example I focus on conditional inference tree, which incorporates tree-structured regression models into conditional inference procedures. While growing a single tree is subject to small changes in the training data, random forests procedure is introduced to address this problem. The sources of diversity for random forests come from the random sampling and restricted set of input variables to be selected. Finally, I introduce R functions to perform model based recursive partitioning. This method incorporates recursive partitioning into conventional parametric model building.

Estimation of retinal vessel caliber using model fitting and random forests

NASA Astrophysics Data System (ADS)

Araújo, Teresa; Mendonça, Ana Maria; Campilho, Aurélio

2017-03-01

Retinal vessel caliber changes are associated with several major diseases, such as diabetes and hypertension. These caliber changes can be evaluated using eye fundus images. However, the clinical assessment is tiresome and prone to errors, motivating the development of automatic methods. An automatic method based on vessel crosssection intensity profile model fitting for the estimation of vessel caliber in retinal images is herein proposed. First, vessels are segmented from the image, vessel centerlines are detected and individual segments are extracted and smoothed. Intensity profiles are extracted perpendicularly to the vessel, and the profile lengths are determined. Then, model fitting is applied to the smoothed profiles. A novel parametric model (DoG-L7) is used, consisting on a Difference-of-Gaussians multiplied by a line which is able to describe profile asymmetry. Finally, the parameters of the best-fit model are used for determining the vessel width through regression using ensembles of bagged regression trees with random sampling of the predictors (random forests). The method is evaluated on the REVIEW public dataset. A precision close to the observers is achieved, outperforming other state-of-the-art methods. The method is robust and reliable for width estimation in images with pathologies and artifacts, with performance independent of the range of diameters.

Genetic analyses of stillbirth in relation to litter size using random regression models.

PubMed

Chen, C Y; Misztal, I; Tsuruta, S; Herring, W O; Holl, J; Culbertson, M

2010-12-01

Estimates of genetic parameters for number of stillborns (NSB) in relation to litter size (LS) were obtained with random regression models (RRM). Data were collected from 4 purebred Duroc nucleus farms between 2004 and 2008. Two data sets with 6,575 litters for the first parity (P1) and 6,259 litters for the second to fifth parity (P2-5) with a total of 8,217 and 5,066 animals in the pedigree were analyzed separately. Number of stillborns was studied as a trait on sow level. Fixed effects were contemporary groups (farm-year-season) and fixed cubic regression coefficients on LS with Legendre polynomials. Models for P2-5 included the fixed effect of parity. Random effects were additive genetic effects for both data sets with permanent environmental effects included for P2-5. Random effects modeled with Legendre polynomials (RRM-L), linear splines (RRM-S), and degree 0 B-splines (RRM-BS) with regressions on LS were used. For P1, the order of polynomial, the number of knots, and the number of intervals used for respective models were quadratic, 3, and 3, respectively. For P2-5, the same parameters were linear, 2, and 2, respectively. Heterogeneous residual variances were considered in the models. For P1, estimates of heritability were 12 to 15%, 5 to 6%, and 6 to 7% in LS 5, 9, and 13, respectively. For P2-5, estimates were 15 to 17%, 4 to 5%, and 4 to 6% in LS 6, 9, and 12, respectively. For P1, average estimates of genetic correlations between LS 5 to 9, 5 to 13, and 9 to 13 were 0.53, -0.29, and 0.65, respectively. For P2-5, same estimates averaged for RRM-L and RRM-S were 0.75, -0.21, and 0.50, respectively. For RRM-BS with 2 intervals, the correlation was 0.66 between LS 5 to 7 and 8 to 13. Parameters obtained by 3 RRM revealed the nonlinear relationship between additive genetic effect of NSB and the environmental deviation of LS. The negative correlations between the 2 extreme LS might possibly indicate different genetic bases on incidence of stillbirth.

Effect of motivational interviewing on rates of early childhood caries: a randomized trial.

PubMed

Harrison, Rosamund; Benton, Tonya; Everson-Stewart, Siobhan; Weinstein, Phil

2007-01-01

The purposes of this randomized controlled trial were to: (1) test motivational interviewing (MI) to prevent early childhood caries; and (2) use Poisson regression for data analysis. A total of 240 South Asian children 6 to 18 months old were enrolled and randomly assigned to either the MI or control condition. Children had a dental exam, and their mothers completed pretested instruments at baseline and 1 and 2 years postintervention. Other covariates that might explain outcomes over and above treatment differences were modeled using Poisson regression. Hazard ratios were produced. Analyses included all participants whenever possible. Poisson regression supported a protective effect of MI (hazard ratio [HR]=0.54 (95%CI=035-0.84)-that is, the M/ group had about a 46% lower rate of dmfs at 2 years than did control children. Similar treatment effect estimates were obtained from models that included, as alternative outcomes, ds, dms, and dmfs, including "white spot lesions." Exploratory analyses revealed that rates of dmfs were higher in children whose mothers had: (1) prechewed their food; (2) been raised in a rural environment; and (3) a higher family income (P<.05). A motivational interviewing-style intervention shows promise to promote preventive behaviors in mothers of young children at high risk for caries.

Large Unbalanced Credit Scoring Using Lasso-Logistic Regression Ensemble

PubMed Central

Wang, Hong; Xu, Qingsong; Zhou, Lifeng

2015-01-01

Recently, various ensemble learning methods with different base classifiers have been proposed for credit scoring problems. However, for various reasons, there has been little research using logistic regression as the base classifier. In this paper, given large unbalanced data, we consider the plausibility of ensemble learning using regularized logistic regression as the base classifier to deal with credit scoring problems. In this research, the data is first balanced and diversified by clustering and bagging algorithms. Then we apply a Lasso-logistic regression learning ensemble to evaluate the credit risks. We show that the proposed algorithm outperforms popular credit scoring models such as decision tree, Lasso-logistic regression and random forests in terms of AUC and F-measure. We also provide two importance measures for the proposed model to identify important variables in the data. PMID:25706988

Restricted spatial regression in practice: Geostatistical models, confounding, and robustness under model misspecification

USGS Publications Warehouse

Hanks, Ephraim M.; Schliep, Erin M.; Hooten, Mevin B.; Hoeting, Jennifer A.

2015-01-01

In spatial generalized linear mixed models (SGLMMs), covariates that are spatially smooth are often collinear with spatially smooth random effects. This phenomenon is known as spatial confounding and has been studied primarily in the case where the spatial support of the process being studied is discrete (e.g., areal spatial data). In this case, the most common approach suggested is restricted spatial regression (RSR) in which the spatial random effects are constrained to be orthogonal to the fixed effects. We consider spatial confounding and RSR in the geostatistical (continuous spatial support) setting. We show that RSR provides computational benefits relative to the confounded SGLMM, but that Bayesian credible intervals under RSR can be inappropriately narrow under model misspecification. We propose a posterior predictive approach to alleviating this potential problem and discuss the appropriateness of RSR in a variety of situations. We illustrate RSR and SGLMM approaches through simulation studies and an analysis of malaria frequencies in The Gambia, Africa.

Exploring the use of random regression models with legendre polynomials to analyze measures of volume of ejaculate in Holstein bulls.

PubMed

Carabaño, M J; Díaz, C; Ugarte, C; Serrano, M

2007-02-01

Artificial insemination centers routinely collect records of quantity and quality of semen of bulls throughout the animals' productive period. The goal of this paper was to explore the use of random regression models with orthogonal polynomials to analyze repeated measures of semen production of Spanish Holstein bulls. A total of 8,773 records of volume of first ejaculate (VFE) collected between 12 and 30 mo of age from 213 Spanish Holstein bulls was analyzed under alternative random regression models. Legendre polynomial functions of increasing order (0 to 6) were fitted to the average trajectory, additive genetic and permanent environmental effects. Age at collection and days in production were used as time variables. Heterogeneous and homogeneous residual variances were alternatively assumed. Analyses were carried out within a Bayesian framework. The logarithm of the marginal density and the cross-validation predictive ability of the data were used as model comparison criteria. Based on both criteria, age at collection as a time variable and heterogeneous residuals models are recommended to analyze changes of VFE over time. Both criteria indicated that fitting random curves for genetic and permanent environmental components as well as for the average trajector improved the quality of models. Furthermore, models with a higher order polynomial for the permanent environmental (5 to 6) than for the genetic components (4 to 5) and the average trajectory (2 to 3) tended to perform best. High-order polynomials were needed to accommodate the highly oscillating nature of the phenotypic values. Heritability and repeatability estimates, disregarding the extremes of the studied period, ranged from 0.15 to 0.35 and from 0.20 to 0.50, respectively, indicating that selection for VFE may be effective at any stage. Small differences among models were observed. Apart from the extremes, estimated correlations between ages decreased steadily from 0.9 and 0.4 for measures 1 mo apart to 0.4 and 0.2 for most distant measures for additive genetic and phenotypic components, respectively. Further investigation to account for environmental factors that may be responsible for the oscillating observations of VFE is needed.

Predicting longitudinal trajectories of health probabilities with random-effects multinomial logit regression.

PubMed

Liu, Xian; Engel, Charles C

2012-12-20

Researchers often encounter longitudinal health data characterized with three or more ordinal or nominal categories. Random-effects multinomial logit models are generally applied to account for potential lack of independence inherent in such clustered data. When parameter estimates are used to describe longitudinal processes, however, random effects, both between and within individuals, need to be retransformed for correctly predicting outcome probabilities. This study attempts to go beyond existing work by developing a retransformation method that derives longitudinal growth trajectories of unbiased health probabilities. We estimated variances of the predicted probabilities by using the delta method. Additionally, we transformed the covariates' regression coefficients on the multinomial logit function, not substantively meaningful, to the conditional effects on the predicted probabilities. The empirical illustration uses the longitudinal data from the Asset and Health Dynamics among the Oldest Old. Our analysis compared three sets of the predicted probabilities of three health states at six time points, obtained from, respectively, the retransformation method, the best linear unbiased prediction, and the fixed-effects approach. The results demonstrate that neglect of retransforming random errors in the random-effects multinomial logit model results in severely biased longitudinal trajectories of health probabilities as well as overestimated effects of covariates on the probabilities. Copyright © 2012 John Wiley & Sons, Ltd.

Ensemble habitat mapping of invasive plant species

USGS Publications Warehouse

Stohlgren, T.J.; Ma, P.; Kumar, S.; Rocca, M.; Morisette, J.T.; Jarnevich, C.S.; Benson, N.

2010-01-01

Ensemble species distribution models combine the strengths of several species environmental matching models, while minimizing the weakness of any one model. Ensemble models may be particularly useful in risk analysis of recently arrived, harmful invasive species because species may not yet have spread to all suitable habitats, leaving species-environment relationships difficult to determine. We tested five individual models (logistic regression, boosted regression trees, random forest, multivariate adaptive regression splines (MARS), and maximum entropy model or Maxent) and ensemble modeling for selected nonnative plant species in Yellowstone and Grand Teton National Parks, Wyoming; Sequoia and Kings Canyon National Parks, California, and areas of interior Alaska. The models are based on field data provided by the park staffs, combined with topographic, climatic, and vegetation predictors derived from satellite data. For the four invasive plant species tested, ensemble models were the only models that ranked in the top three models for both field validation and test data. Ensemble models may be more robust than individual species-environment matching models for risk analysis. ?? 2010 Society for Risk Analysis.

Genetic analysis of partial egg production records in Japanese quail using random regression models.

PubMed

Abou Khadiga, G; Mahmoud, B Y F; Farahat, G S; Emam, A M; El-Full, E A

2017-08-01

The main objectives of this study were to detect the most appropriate random regression model (RRM) to fit the data of monthly egg production in 2 lines (selected and control) of Japanese quail and to test the consistency of different criteria of model choice. Data from 1,200 female Japanese quails for the first 5 months of egg production from 4 consecutive generations of an egg line selected for egg production in the first month (EP1) was analyzed. Eight RRMs with different orders of Legendre polynomials were compared to determine the proper model for analysis. All criteria of model choice suggested that the adequate model included the second-order Legendre polynomials for fixed effects, and the third-order for additive genetic effects and permanent environmental effects. Predictive ability of the best model was the highest among all models (ρ = 0.987). According to the best model fitted to the data, estimates of heritability were relatively low to moderate (0.10 to 0.17) showed a descending pattern from the first to the fifth month of production. A similar pattern was observed for permanent environmental effects with greater estimates in the first (0.36) and second (0.23) months of production than heritability estimates. Genetic correlations between separate production periods were higher (0.18 to 0.93) than their phenotypic counterparts (0.15 to 0.87). The superiority of the selected line over the control was observed through significant (P < 0.05) linear contrast estimates. Significant (P < 0.05) estimates of covariate effect (age at sexual maturity) showed a decreased pattern with greater impact on egg production in earlier ages (first and second months) than later ones. A methodology based on random regression animal models can be recommended for genetic evaluation of egg production in Japanese quail. © 2017 Poultry Science Association Inc.

Methods for estimating population density in data-limited areas: evaluating regression and tree-based models in Peru.

PubMed

Anderson, Weston; Guikema, Seth; Zaitchik, Ben; Pan, William

2014-01-01

Obtaining accurate small area estimates of population is essential for policy and health planning but is often difficult in countries with limited data. In lieu of available population data, small area estimate models draw information from previous time periods or from similar areas. This study focuses on model-based methods for estimating population when no direct samples are available in the area of interest. To explore the efficacy of tree-based models for estimating population density, we compare six different model structures including Random Forest and Bayesian Additive Regression Trees. Results demonstrate that without information from prior time periods, non-parametric tree-based models produced more accurate predictions than did conventional regression methods. Improving estimates of population density in non-sampled areas is important for regions with incomplete census data and has implications for economic, health and development policies.

Methods for Estimating Population Density in Data-Limited Areas: Evaluating Regression and Tree-Based Models in Peru

PubMed Central

Anderson, Weston; Guikema, Seth; Zaitchik, Ben; Pan, William

2014-01-01

Obtaining accurate small area estimates of population is essential for policy and health planning but is often difficult in countries with limited data. In lieu of available population data, small area estimate models draw information from previous time periods or from similar areas. This study focuses on model-based methods for estimating population when no direct samples are available in the area of interest. To explore the efficacy of tree-based models for estimating population density, we compare six different model structures including Random Forest and Bayesian Additive Regression Trees. Results demonstrate that without information from prior time periods, non-parametric tree-based models produced more accurate predictions than did conventional regression methods. Improving estimates of population density in non-sampled areas is important for regions with incomplete census data and has implications for economic, health and development policies. PMID:24992657

Isolating the cow-specific part of residual energy intake in lactating dairy cows using random regressions.

PubMed

Fischer, A; Friggens, N C; Berry, D P; Faverdin, P

2018-07-01

The ability to properly assess and accurately phenotype true differences in feed efficiency among dairy cows is key to the development of breeding programs for improving feed efficiency. The variability among individuals in feed efficiency is commonly characterised by the residual intake approach. Residual feed intake is represented by the residuals of a linear regression of intake on the corresponding quantities of the biological functions that consume (or release) energy. However, the residuals include both, model fitting and measurement errors as well as any variability in cow efficiency. The objective of this study was to isolate the individual animal variability in feed efficiency from the residual component. Two separate models were fitted, in one the standard residual energy intake (REI) was calculated as the residual of a multiple linear regression of lactation average net energy intake (NEI) on lactation average milk energy output, average metabolic BW, as well as lactation loss and gain of body condition score. In the other, a linear mixed model was used to simultaneously fit fixed linear regressions and random cow levels on the biological traits and intercept using fortnight repeated measures for the variables. This method split the predicted NEI in two parts: one quantifying the population mean intercept and coefficients, and one quantifying cow-specific deviations in the intercept and coefficients. The cow-specific part of predicted NEI was assumed to isolate true differences in feed efficiency among cows. NEI and associated energy expenditure phenotypes were available for the first 17 fortnights of lactation from 119 Holstein cows; all fed a constant energy-rich diet. Mixed models fitting cow-specific intercept and coefficients to different combinations of the aforementioned energy expenditure traits, calculated on a fortnightly basis, were compared. The variance of REI estimated with the lactation average model represented only 8% of the variance of measured NEI. Among all compared mixed models, the variance of the cow-specific part of predicted NEI represented between 53% and 59% of the variance of REI estimated from the lactation average model or between 4% and 5% of the variance of measured NEI. The remaining 41% to 47% of the variance of REI estimated with the lactation average model may therefore reflect model fitting errors or measurement errors. In conclusion, the use of a mixed model framework with cow-specific random regressions seems to be a promising method to isolate the cow-specific component of REI in dairy cows.

Parameter estimation of multivariate multiple regression model using bayesian with non-informative Jeffreys’ prior distribution

NASA Astrophysics Data System (ADS)

Saputro, D. R. S.; Amalia, F.; Widyaningsih, P.; Affan, R. C.

2018-05-01

Bayesian method is a method that can be used to estimate the parameters of multivariate multiple regression model. Bayesian method has two distributions, there are prior and posterior distributions. Posterior distribution is influenced by the selection of prior distribution. Jeffreys’ prior distribution is a kind of Non-informative prior distribution. This prior is used when the information about parameter not available. Non-informative Jeffreys’ prior distribution is combined with the sample information resulting the posterior distribution. Posterior distribution is used to estimate the parameter. The purposes of this research is to estimate the parameters of multivariate regression model using Bayesian method with Non-informative Jeffreys’ prior distribution. Based on the results and discussion, parameter estimation of β and Σ which were obtained from expected value of random variable of marginal posterior distribution function. The marginal posterior distributions for β and Σ are multivariate normal and inverse Wishart. However, in calculation of the expected value involving integral of a function which difficult to determine the value. Therefore, approach is needed by generating of random samples according to the posterior distribution characteristics of each parameter using Markov chain Monte Carlo (MCMC) Gibbs sampling algorithm.

Regression-based adaptive sparse polynomial dimensional decomposition for sensitivity analysis

NASA Astrophysics Data System (ADS)

Tang, Kunkun; Congedo, Pietro; Abgrall, Remi

2014-11-01

Polynomial dimensional decomposition (PDD) is employed in this work for global sensitivity analysis and uncertainty quantification of stochastic systems subject to a large number of random input variables. Due to the intimate structure between PDD and Analysis-of-Variance, PDD is able to provide simpler and more direct evaluation of the Sobol' sensitivity indices, when compared to polynomial chaos (PC). Unfortunately, the number of PDD terms grows exponentially with respect to the size of the input random vector, which makes the computational cost of the standard method unaffordable for real engineering applications. In order to address this problem of curse of dimensionality, this work proposes a variance-based adaptive strategy aiming to build a cheap meta-model by sparse-PDD with PDD coefficients computed by regression. During this adaptive procedure, the model representation by PDD only contains few terms, so that the cost to resolve repeatedly the linear system of the least-square regression problem is negligible. The size of the final sparse-PDD representation is much smaller than the full PDD, since only significant terms are eventually retained. Consequently, a much less number of calls to the deterministic model is required to compute the final PDD coefficients.

Random effects coefficient of determination for mixed and meta-analysis models.

PubMed

Demidenko, Eugene; Sargent, James; Onega, Tracy

2012-01-01

The key feature of a mixed model is the presence of random effects. We have developed a coefficient, called the random effects coefficient of determination, [Formula: see text], that estimates the proportion of the conditional variance of the dependent variable explained by random effects. This coefficient takes values from 0 to 1 and indicates how strong the random effects are. The difference from the earlier suggested fixed effects coefficient of determination is emphasized. If [Formula: see text] is close to 0, there is weak support for random effects in the model because the reduction of the variance of the dependent variable due to random effects is small; consequently, random effects may be ignored and the model simplifies to standard linear regression. The value of [Formula: see text] apart from 0 indicates the evidence of the variance reduction in support of the mixed model. If random effects coefficient of determination is close to 1 the variance of random effects is very large and random effects turn into free fixed effects-the model can be estimated using the dummy variable approach. We derive explicit formulas for [Formula: see text] in three special cases: the random intercept model, the growth curve model, and meta-analysis model. Theoretical results are illustrated with three mixed model examples: (1) travel time to the nearest cancer center for women with breast cancer in the U.S., (2) cumulative time watching alcohol related scenes in movies among young U.S. teens, as a risk factor for early drinking onset, and (3) the classic example of the meta-analysis model for combination of 13 studies on tuberculosis vaccine.

Polynomial order selection in random regression models via penalizing adaptively the likelihood.

PubMed

Corrales, J D; Munilla, S; Cantet, R J C

2015-08-01

Orthogonal Legendre polynomials (LP) are used to model the shape of additive genetic and permanent environmental effects in random regression models (RRM). Frequently, the Akaike (AIC) and the Bayesian (BIC) information criteria are employed to select LP order. However, it has been theoretically shown that neither AIC nor BIC is simultaneously optimal in terms of consistency and efficiency. Thus, the goal was to introduce a method, 'penalizing adaptively the likelihood' (PAL), as a criterion to select LP order in RRM. Four simulated data sets and real data (60,513 records, 6675 Colombian Holstein cows) were employed. Nested models were fitted to the data, and AIC, BIC and PAL were calculated for all of them. Results showed that PAL and BIC identified with probability of one the true LP order for the additive genetic and permanent environmental effects, but AIC tended to favour over parameterized models. Conversely, when the true model was unknown, PAL selected the best model with higher probability than AIC. In the latter case, BIC never favoured the best model. To summarize, PAL selected a correct model order regardless of whether the 'true' model was within the set of candidates. © 2015 Blackwell Verlag GmbH.

Personalized Risk Prediction in Clinical Oncology Research: Applications and Practical Issues Using Survival Trees and Random Forests.

PubMed

Hu, Chen; Steingrimsson, Jon Arni

2018-01-01

A crucial component of making individualized treatment decisions is to accurately predict each patient's disease risk. In clinical oncology, disease risks are often measured through time-to-event data, such as overall survival and progression/recurrence-free survival, and are often subject to censoring. Risk prediction models based on recursive partitioning methods are becoming increasingly popular largely due to their ability to handle nonlinear relationships, higher-order interactions, and/or high-dimensional covariates. The most popular recursive partitioning methods are versions of the Classification and Regression Tree (CART) algorithm, which builds a simple interpretable tree structured model. With the aim of increasing prediction accuracy, the random forest algorithm averages multiple CART trees, creating a flexible risk prediction model. Risk prediction models used in clinical oncology commonly use both traditional demographic and tumor pathological factors as well as high-dimensional genetic markers and treatment parameters from multimodality treatments. In this article, we describe the most commonly used extensions of the CART and random forest algorithms to right-censored outcomes. We focus on how they differ from the methods for noncensored outcomes, and how the different splitting rules and methods for cost-complexity pruning impact these algorithms. We demonstrate these algorithms by analyzing a randomized Phase III clinical trial of breast cancer. We also conduct Monte Carlo simulations to compare the prediction accuracy of survival forests with more commonly used regression models under various scenarios. These simulation studies aim to evaluate how sensitive the prediction accuracy is to the underlying model specifications, the choice of tuning parameters, and the degrees of missing covariates.

Effect of folic acid on appetite in children: ordinal logistic and fuzzy logistic regressions.

PubMed

Namdari, Mahshid; Abadi, Alireza; Taheri, S Mahmoud; Rezaei, Mansour; Kalantari, Naser; Omidvar, Nasrin

2014-03-01

Reduced appetite and low food intake are often a concern in preschool children, since it can lead to malnutrition, a leading cause of impaired growth and mortality in childhood. It is occasionally considered that folic acid has a positive effect on appetite enhancement and consequently growth in children. The aim of this study was to assess the effect of folic acid on the appetite of preschool children 3 to 6 y old. The study sample included 127 children ages 3 to 6 who were randomly selected from 20 preschools in the city of Tehran in 2011. Since appetite was measured by linguistic terms, a fuzzy logistic regression was applied for modeling. The obtained results were compared with a statistical ordinal logistic model. After controlling for the potential confounders, in a statistical ordinal logistic model, serum folate showed a significantly positive effect on appetite. A small but positive effect of folate was detected by fuzzy logistic regression. Based on fuzzy regression, the risk for poor appetite in preschool children was related to the employment status of their mothers. In this study, a positive association was detected between the levels of serum folate and improved appetite. For further investigation, a randomized controlled, double-blind clinical trial could be helpful to address causality. Copyright © 2014 Elsevier Inc. All rights reserved.

Genetic correlations between body condition scores and fertility in dairy cattle using bivariate random regression models.

PubMed

De Haas, Y; Janss, L L G; Kadarmideen, H N

2007-10-01

Genetic correlations between body condition score (BCS) and fertility traits in dairy cattle were estimated using bivariate random regression models. BCS was recorded by the Swiss Holstein Association on 22,075 lactating heifers (primiparous cows) from 856 sires. Fertility data during first lactation were extracted for 40,736 cows. The fertility traits were days to first service (DFS), days between first and last insemination (DFLI), calving interval (CI), number of services per conception (NSPC) and conception rate to first insemination (CRFI). A bivariate model was used to estimate genetic correlations between BCS as a longitudinal trait by random regression components, and daughter's fertility at the sire level as a single lactation measurement. Heritability of BCS was 0.17, and heritabilities for fertility traits were low (0.01-0.08). Genetic correlations between BCS and fertility over the lactation varied from: -0.45 to -0.14 for DFS; -0.75 to 0.03 for DFLI; from -0.59 to -0.02 for CI; from -0.47 to 0.33 for NSPC and from 0.08 to 0.82 for CRFI. These results show (genetic) interactions between fat reserves and reproduction along the lactation trajectory of modern dairy cows, which can be useful in genetic selection as well as in management. Maximum genetic gain in fertility from indirect selection on BCS should be based on measurements taken in mid lactation when the genetic variance for BCS is largest, and the genetic correlations between BCS and fertility is strongest.

Using Structured Additive Regression Models to Estimate Risk Factors of Malaria: Analysis of 2010 Malawi Malaria Indicator Survey Data

PubMed Central

Chirombo, James; Lowe, Rachel; Kazembe, Lawrence

2014-01-01

Background After years of implementing Roll Back Malaria (RBM) interventions, the changing landscape of malaria in terms of risk factors and spatial pattern has not been fully investigated. This paper uses the 2010 malaria indicator survey data to investigate if known malaria risk factors remain relevant after many years of interventions. Methods We adopted a structured additive logistic regression model that allowed for spatial correlation, to more realistically estimate malaria risk factors. Our model included child and household level covariates, as well as climatic and environmental factors. Continuous variables were modelled by assuming second order random walk priors, while spatial correlation was specified as a Markov random field prior, with fixed effects assigned diffuse priors. Inference was fully Bayesian resulting in an under five malaria risk map for Malawi. Results Malaria risk increased with increasing age of the child. With respect to socio-economic factors, the greater the household wealth, the lower the malaria prevalence. A general decline in malaria risk was observed as altitude increased. Minimum temperatures and average total rainfall in the three months preceding the survey did not show a strong association with disease risk. Conclusions The structured additive regression model offered a flexible extension to standard regression models by enabling simultaneous modelling of possible nonlinear effects of continuous covariates, spatial correlation and heterogeneity, while estimating usual fixed effects of categorical and continuous observed variables. Our results confirmed that malaria epidemiology is a complex interaction of biotic and abiotic factors, both at the individual, household and community level and that risk factors are still relevant many years after extensive implementation of RBM activities. PMID:24991915

Using structured additive regression models to estimate risk factors of malaria: analysis of 2010 Malawi malaria indicator survey data.

PubMed

Chirombo, James; Lowe, Rachel; Kazembe, Lawrence

2014-01-01

After years of implementing Roll Back Malaria (RBM) interventions, the changing landscape of malaria in terms of risk factors and spatial pattern has not been fully investigated. This paper uses the 2010 malaria indicator survey data to investigate if known malaria risk factors remain relevant after many years of interventions. We adopted a structured additive logistic regression model that allowed for spatial correlation, to more realistically estimate malaria risk factors. Our model included child and household level covariates, as well as climatic and environmental factors. Continuous variables were modelled by assuming second order random walk priors, while spatial correlation was specified as a Markov random field prior, with fixed effects assigned diffuse priors. Inference was fully Bayesian resulting in an under five malaria risk map for Malawi. Malaria risk increased with increasing age of the child. With respect to socio-economic factors, the greater the household wealth, the lower the malaria prevalence. A general decline in malaria risk was observed as altitude increased. Minimum temperatures and average total rainfall in the three months preceding the survey did not show a strong association with disease risk. The structured additive regression model offered a flexible extension to standard regression models by enabling simultaneous modelling of possible nonlinear effects of continuous covariates, spatial correlation and heterogeneity, while estimating usual fixed effects of categorical and continuous observed variables. Our results confirmed that malaria epidemiology is a complex interaction of biotic and abiotic factors, both at the individual, household and community level and that risk factors are still relevant many years after extensive implementation of RBM activities.

Small-Sample Adjustments for Tests of Moderators and Model Fit in Robust Variance Estimation in Meta-Regression

ERIC Educational Resources Information Center

Tipton, Elizabeth; Pustejovsky, James E.

2015-01-01

Randomized experiments are commonly used to evaluate the effectiveness of educational interventions. The goal of the present investigation is to develop small-sample corrections for multiple contrast hypothesis tests (i.e., F-tests) such as the omnibus test of meta-regression fit or a test for equality of three or more levels of a categorical…

Solving large mixed linear models using preconditioned conjugate gradient iteration.

PubMed

Strandén, I; Lidauer, M

1999-12-01

Continuous evaluation of dairy cattle with a random regression test-day model requires a fast solving method and algorithm. A new computing technique feasible in Jacobi and conjugate gradient based iterative methods using iteration on data is presented. In the new computing technique, the calculations in multiplication of a vector by a matrix were recorded to three steps instead of the commonly used two steps. The three-step method was implemented in a general mixed linear model program that used preconditioned conjugate gradient iteration. Performance of this program in comparison to other general solving programs was assessed via estimation of breeding values using univariate, multivariate, and random regression test-day models. Central processing unit time per iteration with the new three-step technique was, at best, one-third that needed with the old technique. Performance was best with the test-day model, which was the largest and most complex model used. The new program did well in comparison to other general software. Programs keeping the mixed model equations in random access memory required at least 20 and 435% more time to solve the univariate and multivariate animal models, respectively. Computations of the second best iteration on data took approximately three and five times longer for the animal and test-day models, respectively, than did the new program. Good performance was due to fast computing time per iteration and quick convergence to the final solutions. Use of preconditioned conjugate gradient based methods in solving large breeding value problems is supported by our findings.

The regression discontinuity design showed to be a valid alternative to a randomized controlled trial for estimating treatment effects.

PubMed

Maas, Iris L; Nolte, Sandra; Walter, Otto B; Berger, Thomas; Hautzinger, Martin; Hohagen, Fritz; Lutz, Wolfgang; Meyer, Björn; Schröder, Johanna; Späth, Christina; Klein, Jan Philipp; Moritz, Steffen; Rose, Matthias

2017-02-01

To compare treatment effect estimates obtained from a regression discontinuity (RD) design with results from an actual randomized controlled trial (RCT). Data from an RCT (EVIDENT), which studied the effect of an Internet intervention on depressive symptoms measured with the Patient Health Questionnaire (PHQ-9), were used to perform an RD analysis, in which treatment allocation was determined by a cutoff value at baseline (PHQ-9 = 10). A linear regression model was fitted to the data, selecting participants above the cutoff who had received the intervention (n = 317) and control participants below the cutoff (n = 187). Outcome was PHQ-9 sum score 12 weeks after baseline. Robustness of the effect estimate was studied; the estimate was compared with the RCT treatment effect. The final regression model showed a regression coefficient of -2.29 [95% confidence interval (CI): -3.72 to -.85] compared with a treatment effect found in the RCT of -1.57 (95% CI: -2.07 to -1.07). Although the estimates obtained from two designs are not equal, their confidence intervals overlap, suggesting that an RD design can be a valid alternative for RCTs. This finding is particularly important for situations where an RCT may not be feasible or ethical as is often the case in clinical research settings. Copyright © 2016 Elsevier Inc. All rights reserved.

Benchmarking dairy herd health status using routinely recorded herd summary data.

PubMed

Parker Gaddis, K L; Cole, J B; Clay, J S; Maltecca, C

2016-02-01

Genetic improvement of dairy cattle health through the use of producer-recorded data has been determined to be feasible. Low estimated heritabilities indicate that genetic progress will be slow. Variation observed in lowly heritable traits can largely be attributed to nongenetic factors, such as the environment. More rapid improvement of dairy cattle health may be attainable if herd health programs incorporate environmental and managerial aspects. More than 1,100 herd characteristics are regularly recorded on farm test-days. We combined these data with producer-recorded health event data, and parametric and nonparametric models were used to benchmark herd and cow health status. Health events were grouped into 3 categories for analyses: mastitis, reproductive, and metabolic. Both herd incidence and individual incidence were used as dependent variables. Models implemented included stepwise logistic regression, support vector machines, and random forests. At both the herd and individual levels, random forest models attained the highest accuracy for predicting health status in all health event categories when evaluated with 10-fold cross-validation. Accuracy (SD) ranged from 0.61 (0.04) to 0.63 (0.04) when using random forest models at the herd level. Accuracy of prediction (SD) at the individual cow level ranged from 0.87 (0.06) to 0.93 (0.001) with random forest models. Highly significant variables and key words from logistic regression and random forest models were also investigated. All models identified several of the same key factors for each health event category, including movement out of the herd, size of the herd, and weather-related variables. We concluded that benchmarking health status using routinely collected herd data is feasible. Nonparametric models were better suited to handle this complex data with numerous variables. These data mining techniques were able to perform prediction of health status and could add evidence to personal experience in herd management. Copyright © 2016 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

Predicting recreational water quality advisories: A comparison of statistical methods

USGS Publications Warehouse

Brooks, Wesley R.; Corsi, Steven R.; Fienen, Michael N.; Carvin, Rebecca B.

2016-01-01

Epidemiological studies indicate that fecal indicator bacteria (FIB) in beach water are associated with illnesses among people having contact with the water. In order to mitigate public health impacts, many beaches are posted with an advisory when the concentration of FIB exceeds a beach action value. The most commonly used method of measuring FIB concentration takes 18–24 h before returning a result. In order to avoid the 24 h lag, it has become common to ”nowcast” the FIB concentration using statistical regressions on environmental surrogate variables. Most commonly, nowcast models are estimated using ordinary least squares regression, but other regression methods from the statistical and machine learning literature are sometimes used. This study compares 14 regression methods across 7 Wisconsin beaches to identify which consistently produces the most accurate predictions. A random forest model is identified as the most accurate, followed by multiple regression fit using the adaptive LASSO.

MIXREG: a computer program for mixed-effects regression analysis with autocorrelated errors.

PubMed

Hedeker, D; Gibbons, R D

1996-05-01

MIXREG is a program that provides estimates for a mixed-effects regression model (MRM) for normally-distributed response data including autocorrelated errors. This model can be used for analysis of unbalanced longitudinal data, where individuals may be measured at a different number of timepoints, or even at different timepoints. Autocorrelated errors of a general form or following an AR(1), MA(1), or ARMA(1,1) form are allowable. This model can also be used for analysis of clustered data, where the mixed-effects model assumes data within clusters are dependent. The degree of dependency is estimated jointly with estimates of the usual model parameters, thus adjusting for clustering. MIXREG uses maximum marginal likelihood estimation, utilizing both the EM algorithm and a Fisher-scoring solution. For the scoring solution, the covariance matrix of the random effects is expressed in its Gaussian decomposition, and the diagonal matrix reparameterized using the exponential transformation. Estimation of the individual random effects is accomplished using an empirical Bayes approach. Examples illustrating usage and features of MIXREG are provided.

wayGoo recommender system: personalized recommendations for events scheduling, based on static and real-time information

NASA Astrophysics Data System (ADS)

Thanos, Konstantinos-Georgios; Thomopoulos, Stelios C. A.

2016-05-01

wayGoo is a fully functional application whose main functionalities include content geolocation, event scheduling, and indoor navigation. However, significant information about events do not reach users' attention, either because of the size of this information or because some information comes from real - time data sources. The purpose of this work is to facilitate event management operations by prioritizing the presented events, based on users' interests using both, static and real - time data. Through the wayGoo interface, users select conceptual topics that are interesting for them. These topics constitute a browsing behavior vector which is used for learning users' interests implicitly, without being intrusive. Then, the system estimates user preferences and return an events list sorted from the most preferred one to the least. User preferences are modeled via a Naïve Bayesian Network which consists of: a) the `decision' random variable corresponding to users' decision on attending an event, b) the `distance' random variable, modeled by a linear regression that estimates the probability that the distance between a user and each event destination is not discouraging, ` the seat availability' random variable, modeled by a linear regression, which estimates the probability that the seat availability is encouraging d) and the `relevance' random variable, modeled by a clustering - based collaborative filtering, which determines the relevance of each event users' interests. Finally, experimental results show that the proposed system contribute essentially to assisting users in browsing and selecting events to attend.

Regression trees for predicting mortality in patients with cardiovascular disease: What improvement is achieved by using ensemble-based methods?

PubMed Central

Austin, Peter C; Lee, Douglas S; Steyerberg, Ewout W; Tu, Jack V

2012-01-01

In biomedical research, the logistic regression model is the most commonly used method for predicting the probability of a binary outcome. While many clinical researchers have expressed an enthusiasm for regression trees, this method may have limited accuracy for predicting health outcomes. We aimed to evaluate the improvement that is achieved by using ensemble-based methods, including bootstrap aggregation (bagging) of regression trees, random forests, and boosted regression trees. We analyzed 30-day mortality in two large cohorts of patients hospitalized with either acute myocardial infarction (N = 16,230) or congestive heart failure (N = 15,848) in two distinct eras (1999–2001 and 2004–2005). We found that both the in-sample and out-of-sample prediction of ensemble methods offered substantial improvement in predicting cardiovascular mortality compared to conventional regression trees. However, conventional logistic regression models that incorporated restricted cubic smoothing splines had even better performance. We conclude that ensemble methods from the data mining and machine learning literature increase the predictive performance of regression trees, but may not lead to clear advantages over conventional logistic regression models for predicting short-term mortality in population-based samples of subjects with cardiovascular disease. PMID:22777999

A method for fitting regression splines with varying polynomial order in the linear mixed model.

PubMed

Edwards, Lloyd J; Stewart, Paul W; MacDougall, James E; Helms, Ronald W

2006-02-15

The linear mixed model has become a widely used tool for longitudinal analysis of continuous variables. The use of regression splines in these models offers the analyst additional flexibility in the formulation of descriptive analyses, exploratory analyses and hypothesis-driven confirmatory analyses. We propose a method for fitting piecewise polynomial regression splines with varying polynomial order in the fixed effects and/or random effects of the linear mixed model. The polynomial segments are explicitly constrained by side conditions for continuity and some smoothness at the points where they join. By using a reparameterization of this explicitly constrained linear mixed model, an implicitly constrained linear mixed model is constructed that simplifies implementation of fixed-knot regression splines. The proposed approach is relatively simple, handles splines in one variable or multiple variables, and can be easily programmed using existing commercial software such as SAS or S-plus. The method is illustrated using two examples: an analysis of longitudinal viral load data from a study of subjects with acute HIV-1 infection and an analysis of 24-hour ambulatory blood pressure profiles.

Simulation of land use change in the three gorges reservoir area based on CART-CA

NASA Astrophysics Data System (ADS)

Yuan, Min

2018-05-01

This study proposes a new method to simulate spatiotemporal complex multiple land uses by using classification and regression tree algorithm (CART) based CA model. In this model, we use classification and regression tree algorithm to calculate land class conversion probability, and combine neighborhood factor, random factor to extract cellular transformation rules. The overall Kappa coefficient is 0.8014 and the overall accuracy is 0.8821 in the land dynamic simulation results of the three gorges reservoir area from 2000 to 2010, and the simulation results are satisfactory.

Genetic evaluation of egg production curve in Thai native chickens by random regression and spline models.

PubMed

Mookprom, S; Boonkum, W; Kunhareang, S; Siripanya, S; Duangjinda, M

2017-02-01

The objective of this research is to investigate appropriate random regression models with various covariance functions, for the genetic evaluation of test-day egg production. Data included 7,884 monthly egg production records from 657 Thai native chickens (Pradu Hang Dam) that were obtained during the first to sixth generation and were born during 2007 to 2014 at the Research and Development Network Center for Animal Breeding (Native Chickens), Khon Kaen University. Average annual and monthly egg productions were 117 ± 41 and 10.20 ± 6.40 eggs, respectively. Nine random regression models were analyzed using the Wilmink function (WM), Koops and Grossman function (KG), Legendre polynomials functions with second, third, and fourth orders (LG2, LG3, LG4), and spline functions with 4, 5, 6, and 8 knots (SP4, SP5, SP6, and SP8). All covariance functions were nested within the same additive genetic and permanent environmental random effects, and the variance components were estimated by Restricted Maximum Likelihood (REML). In model comparisons, mean square error (MSE) and the coefficient of detemination (R 2 ) calculated the goodness of fit; and the correlation between observed and predicted values [Formula: see text] was used to calculate the cross-validated predictive abilities. We found that the covariance functions of SP5, SP6, and SP8 proved appropriate for the genetic evaluation of the egg production curves for Thai native chickens. The estimated heritability of monthly egg production ranged from 0.07 to 0.39, and the highest heritability was found during the first to third months of egg production. In conclusion, the spline functions within monthly egg production can be applied to breeding programs for the improvement of both egg number and persistence of egg production. © 2016 Poultry Science Association Inc.

Influence of therapist competence and quantity of cognitive behavioural therapy on suicidal behaviour and inpatient hospitalisation in a randomised controlled trial in borderline personality disorder: further analyses of treatment effects in the BOSCOT study.

PubMed

Norrie, John; Davidson, Kate; Tata, Philip; Gumley, Andrew

2013-09-01

We investigated the treatment effects reported from a high-quality randomized controlled trial of cognitive behavioural therapy (CBT) for 106 people with borderline personality disorder attending community-based clinics in the UK National Health Service - the BOSCOT trial. Specifically, we examined whether the amount of therapy and therapist competence had an impact on our primary outcome, the number of suicidal acts, using instrumental variables regression modelling. Randomized controlled trial. Participants from across three sites (London, Glasgow, and Ayrshire/Arran) were randomized equally to CBT for personality disorders (CBTpd) plus Treatment as Usual or to Treatment as Usual. Treatment as Usual varied between sites and individuals, but was consistent with routine treatment in the UK National Health Service at the time. CBTpd comprised an average 16 sessions (range 0-35) over 12 months. We used instrumental variable regression modelling to estimate the impact of quantity and quality of therapy received (recording activities and behaviours that took place after randomization) on number of suicidal acts and inpatient psychiatric hospitalization. A total of 101 participants provided full outcome data at 2 years post randomization. The previously reported intention-to-treat (ITT) results showed on average a reduction of 0.91 (95% confidence interval 0.15-1.67) suicidal acts over 2 years for those randomized to CBT. By incorporating the influence of quantity of therapy and therapist competence, we show that this estimate of the effect of CBTpd could be approximately two to three times greater for those receiving the right amount of therapy from a competent therapist. Trials should routinely control for and collect data on both quantity of therapy and therapist competence, which can be used, via instrumental variable regression modelling, to estimate treatment effects for optimal delivery of therapy. Such estimates complement rather than replace the ITT results, which are properly the principal analysis results from such trials. © 2013 The British Psychological Society.

A bioavailable strontium isoscape for Western Europe: A machine learning approach

PubMed Central

von Holstein, Isabella C. C.; Laffoon, Jason E.; Willmes, Malte; Liu, Xiao-Ming; Davies, Gareth R.

2018-01-01

Strontium isotope ratios (87Sr/86Sr) are gaining considerable interest as a geolocation tool and are now widely applied in archaeology, ecology, and forensic research. However, their application for provenance requires the development of baseline models predicting surficial 87Sr/86Sr variations (“isoscapes”). A variety of empirically-based and process-based models have been proposed to build terrestrial 87Sr/86Sr isoscapes but, in their current forms, those models are not mature enough to be integrated with continuous-probability surface models used in geographic assignment. In this study, we aim to overcome those limitations and to predict 87Sr/86Sr variations across Western Europe by combining process-based models and a series of remote-sensing geospatial products into a regression framework. We find that random forest regression significantly outperforms other commonly used regression and interpolation methods, and efficiently predicts the multi-scale patterning of 87Sr/86Sr variations by accounting for geological, geomorphological and atmospheric controls. Random forest regression also provides an easily interpretable and flexible framework to integrate different types of environmental auxiliary variables required to model the multi-scale patterning of 87Sr/86Sr variability. The method is transferable to different scales and resolutions and can be applied to the large collection of geospatial data available at local and global levels. The isoscape generated in this study provides the most accurate 87Sr/86Sr predictions in bioavailable strontium for Western Europe (R2 = 0.58 and RMSE = 0.0023) to date, as well as a conservative estimate of spatial uncertainty by applying quantile regression forest. We anticipate that the method presented in this study combined with the growing numbers of bioavailable 87Sr/86Sr data and satellite geospatial products will extend the applicability of the 87Sr/86Sr geo-profiling tool in provenance applications. PMID:29847595

Analysis of Learning Curve Fitting Techniques.

DTIC Science & Technology

1987-09-01

1986. 15. Neter, John and others. Applied Linear Regression Models. Homewood IL: Irwin, 19-33. 16. SAS User’s Guide: Basics, Version 5 Edition. SAS... Linear Regression Techniques (15:23-52). Random errors are assumed to be normally distributed when using -# ordinary least-squares, according to Johnston...lot estimated by the improvement curve formula. For a more detailed explanation of the ordinary least-squares technique, see Neter, et. al., Applied

Logistic quantile regression provides improved estimates for bounded avian counts: a case study of California Spotted Owl fledgling production

Treesearch

Brian S. Cade; Barry R. Noon; Rick D. Scherer; John J. Keane

2017-01-01

Counts of avian fledglings, nestlings, or clutch size that are bounded below by zero and above by some small integer form a discrete random variable distribution that is not approximated well by conventional parametric count distributions such as the Poisson or negative binomial. We developed a logistic quantile regression model to provide estimates of the empirical...

Inferring genetic parameters of lactation in Tropical Milking Criollo cattle with random regression test-day models.

PubMed

Santellano-Estrada, E; Becerril-Pérez, C M; de Alba, J; Chang, Y M; Gianola, D; Torres-Hernández, G; Ramírez-Valverde, R

2008-11-01

This study inferred genetic and permanent environmental variation of milk yield in Tropical Milking Criollo cattle and compared 5 random regression test-day models using Wilmink's function and Legendre polynomials. Data consisted of 15,377 test-day records from 467 Tropical Milking Criollo cows that calved between 1974 and 2006 in the tropical lowlands of the Gulf Coast of Mexico and in southern Nicaragua. Estimated heritabilities of test-day milk yields ranged from 0.18 to 0.45, and repeatabilities ranged from 0.35 to 0.68 for the period spanning from 6 to 400 d in milk. Genetic correlation between days in milk 10 and 400 was around 0.50 but greater than 0.90 for most pairs of test days. The model that used first-order Legendre polynomials for additive genetic effects and second-order Legendre polynomials for permanent environmental effects gave the smallest residual variance and was also favored by the Akaike information criterion and likelihood ratio tests.

Multi-fidelity Gaussian process regression for prediction of random fields

DOE Office of Scientific and Technical Information (OSTI.GOV)

Parussini, L.; Venturi, D., E-mail: venturi@ucsc.edu; Perdikaris, P.

We propose a new multi-fidelity Gaussian process regression (GPR) approach for prediction of random fields based on observations of surrogate models or hierarchies of surrogate models. Our method builds upon recent work on recursive Bayesian techniques, in particular recursive co-kriging, and extends it to vector-valued fields and various types of covariances, including separable and non-separable ones. The framework we propose is general and can be used to perform uncertainty propagation and quantification in model-based simulations, multi-fidelity data fusion, and surrogate-based optimization. We demonstrate the effectiveness of the proposed recursive GPR techniques through various examples. Specifically, we study the stochastic Burgersmore » equation and the stochastic Oberbeck–Boussinesq equations describing natural convection within a square enclosure. In both cases we find that the standard deviation of the Gaussian predictors as well as the absolute errors relative to benchmark stochastic solutions are very small, suggesting that the proposed multi-fidelity GPR approaches can yield highly accurate results.« less

Egg production forecasting: Determining efficient modeling approaches.

PubMed

Ahmad, H A

2011-12-01

Several mathematical or statistical and artificial intelligence models were developed to compare egg production forecasts in commercial layers. Initial data for these models were collected from a comparative layer trial on commercial strains conducted at the Poultry Research Farms, Auburn University. Simulated data were produced to represent new scenarios by using means and SD of egg production of the 22 commercial strains. From the simulated data, random examples were generated for neural network training and testing for the weekly egg production prediction from wk 22 to 36. Three neural network architectures-back-propagation-3, Ward-5, and the general regression neural network-were compared for their efficiency to forecast egg production, along with other traditional models. The general regression neural network gave the best-fitting line, which almost overlapped with the commercial egg production data, with an R(2) of 0.71. The general regression neural network-predicted curve was compared with original egg production data, the average curves of white-shelled and brown-shelled strains, linear regression predictions, and the Gompertz nonlinear model. The general regression neural network was superior in all these comparisons and may be the model of choice if the initial overprediction is managed efficiently. In general, neural network models are efficient, are easy to use, require fewer data, and are practical under farm management conditions to forecast egg production.

Sample size estimation for alternating logistic regressions analysis of multilevel randomized community trials of under-age drinking.

PubMed

Reboussin, Beth A; Preisser, John S; Song, Eun-Young; Wolfson, Mark

2012-07-01

Under-age drinking is an enormous public health issue in the USA. Evidence that community level structures may impact on under-age drinking has led to a proliferation of efforts to change the environment surrounding the use of alcohol. Although the focus of these efforts is to reduce drinking by individual youths, environmental interventions are typically implemented at the community level with entire communities randomized to the same intervention condition. A distinct feature of these trials is the tendency of the behaviours of individuals residing in the same community to be more alike than that of others residing in different communities, which is herein called 'clustering'. Statistical analyses and sample size calculations must account for this clustering to avoid type I errors and to ensure an appropriately powered trial. Clustering itself may also be of scientific interest. We consider the alternating logistic regressions procedure within the population-averaged modelling framework to estimate the effect of a law enforcement intervention on the prevalence of under-age drinking behaviours while modelling the clustering at multiple levels, e.g. within communities and within neighbourhoods nested within communities, by using pairwise odds ratios. We then derive sample size formulae for estimating intervention effects when planning a post-test-only or repeated cross-sectional community-randomized trial using the alternating logistic regressions procedure.

Modeling Verdict Outcomes Using Social Network Measures: The Watergate and Caviar Network Cases.

PubMed

Masías, Víctor Hugo; Valle, Mauricio; Morselli, Carlo; Crespo, Fernando; Vargas, Augusto; Laengle, Sigifredo

2016-01-01

Modelling criminal trial verdict outcomes using social network measures is an emerging research area in quantitative criminology. Few studies have yet analyzed which of these measures are the most important for verdict modelling or which data classification techniques perform best for this application. To compare the performance of different techniques in classifying members of a criminal network, this article applies three different machine learning classifiers-Logistic Regression, Naïve Bayes and Random Forest-with a range of social network measures and the necessary databases to model the verdicts in two real-world cases: the U.S. Watergate Conspiracy of the 1970's and the now-defunct Canada-based international drug trafficking ring known as the Caviar Network. In both cases it was found that the Random Forest classifier did better than either Logistic Regression or Naïve Bayes, and its superior performance was statistically significant. This being so, Random Forest was used not only for classification but also to assess the importance of the measures. For the Watergate case, the most important one proved to be betweenness centrality while for the Caviar Network, it was the effective size of the network. These results are significant because they show that an approach combining machine learning with social network analysis not only can generate accurate classification models but also helps quantify the importance social network variables in modelling verdict outcomes. We conclude our analysis with a discussion and some suggestions for future work in verdict modelling using social network measures.

Random Regression Models Are Suitable to Substitute the Traditional 305-Day Lactation Model in Genetic Evaluations of Holstein Cattle in Brazil

PubMed Central

Padilha, Alessandro Haiduck; Cobuci, Jaime Araujo; Costa, Cláudio Napolis; Neto, José Braccini

2016-01-01

The aim of this study was to compare two random regression models (RRM) fitted by fourth (RRM4) and fifth-order Legendre polynomials (RRM5) with a lactation model (LM) for evaluating Holstein cattle in Brazil. Two datasets with the same animals were prepared for this study. To apply test-day RRM and LMs, 262,426 test day records and 30,228 lactation records covering 305 days were prepared, respectively. The lowest values of Akaike’s information criterion, Bayesian information criterion, and estimates of the maximum of the likelihood function (−2LogL) were for RRM4. Heritability for 305-day milk yield (305MY) was 0.23 (RRM4), 0.24 (RRM5), and 0.21 (LM). Heritability, additive genetic and permanent environmental variances of test days on days in milk was from 0.16 to 0.27, from 3.76 to 6.88 and from 11.12 to 20.21, respectively. Additive genetic correlations between test days ranged from 0.20 to 0.99. Permanent environmental correlations between test days were between 0.07 and 0.99. Standard deviations of average estimated breeding values (EBVs) for 305MY from RRM4 and RRM5 were from 11% to 30% higher for bulls and around 28% higher for cows than that in LM. Rank correlations between RRM EBVs and LM EBVs were between 0.86 to 0.96 for bulls and 0.80 to 0.87 for cows. Average percentage of gain in reliability of EBVs for 305-day yield increased from 4% to 17% for bulls and from 23% to 24% for cows when reliability of EBVs from RRM models was compared to those from LM model. Random regression model fitted by fourth order Legendre polynomials is recommended for genetic evaluations of Brazilian Holstein cattle because of the higher reliability in the estimation of breeding values. PMID:26954176

Random Regression Models Are Suitable to Substitute the Traditional 305-Day Lactation Model in Genetic Evaluations of Holstein Cattle in Brazil.

PubMed

Padilha, Alessandro Haiduck; Cobuci, Jaime Araujo; Costa, Cláudio Napolis; Neto, José Braccini

2016-06-01

The aim of this study was to compare two random regression models (RRM) fitted by fourth (RRM4) and fifth-order Legendre polynomials (RRM5) with a lactation model (LM) for evaluating Holstein cattle in Brazil. Two datasets with the same animals were prepared for this study. To apply test-day RRM and LMs, 262,426 test day records and 30,228 lactation records covering 305 days were prepared, respectively. The lowest values of Akaike's information criterion, Bayesian information criterion, and estimates of the maximum of the likelihood function (-2LogL) were for RRM4. Heritability for 305-day milk yield (305MY) was 0.23 (RRM4), 0.24 (RRM5), and 0.21 (LM). Heritability, additive genetic and permanent environmental variances of test days on days in milk was from 0.16 to 0.27, from 3.76 to 6.88 and from 11.12 to 20.21, respectively. Additive genetic correlations between test days ranged from 0.20 to 0.99. Permanent environmental correlations between test days were between 0.07 and 0.99. Standard deviations of average estimated breeding values (EBVs) for 305MY from RRM4 and RRM5 were from 11% to 30% higher for bulls and around 28% higher for cows than that in LM. Rank correlations between RRM EBVs and LM EBVs were between 0.86 to 0.96 for bulls and 0.80 to 0.87 for cows. Average percentage of gain in reliability of EBVs for 305-day yield increased from 4% to 17% for bulls and from 23% to 24% for cows when reliability of EBVs from RRM models was compared to those from LM model. Random regression model fitted by fourth order Legendre polynomials is recommended for genetic evaluations of Brazilian Holstein cattle because of the higher reliability in the estimation of breeding values.

Linear regression analysis of survival data with missing censoring indicators.

PubMed

Wang, Qihua; Dinse, Gregg E

2011-04-01

Linear regression analysis has been studied extensively in a random censorship setting, but typically all of the censoring indicators are assumed to be observed. In this paper, we develop synthetic data methods for estimating regression parameters in a linear model when some censoring indicators are missing. We define estimators based on regression calibration, imputation, and inverse probability weighting techniques, and we prove all three estimators are asymptotically normal. The finite-sample performance of each estimator is evaluated via simulation. We illustrate our methods by assessing the effects of sex and age on the time to non-ambulatory progression for patients in a brain cancer clinical trial.

Teacher consultation and coaching within mental health practice: classroom and child effects in urban elementary schools.

PubMed

Cappella, Elise; Hamre, Bridget K; Kim, Ha Yeon; Henry, David B; Frazier, Stacy L; Atkins, Marc S; Schoenwald, Sonja K

2012-08-01

To examine effects of a teacher consultation and coaching program delivered by school and community mental health professionals on change in observed classroom interactions and child functioning across one school year. Thirty-six classrooms within 5 urban elementary schools (87% Latino, 11% Black) were randomly assigned to intervention (training + consultation/coaching) and control (training only) conditions. Classroom and child outcomes (n = 364; 43% girls) were assessed in the fall and spring. Random effects regression models showed main effects of intervention on teacher-student relationship closeness, academic self-concept, and peer victimization. Results of multiple regression models showed levels of observed teacher emotional support in the fall moderated intervention impact on emotional support at the end of the school year. Results suggest teacher consultation and coaching can be integrated within existing mental health activities in urban schools and impact classroom effectiveness and child adaptation across multiple domains. © 2012 American Psychological Association

Multilevel Modeling in Psychosomatic Medicine Research

PubMed Central

Myers, Nicholas D.; Brincks, Ahnalee M.; Ames, Allison J.; Prado, Guillermo J.; Penedo, Frank J.; Benedict, Catherine

2012-01-01

The primary purpose of this manuscript is to provide an overview of multilevel modeling for Psychosomatic Medicine readers and contributors. The manuscript begins with a general introduction to multilevel modeling. Multilevel regression modeling at two-levels is emphasized because of its prevalence in psychosomatic medicine research. Simulated datasets based on some core ideas from the Familias Unidas effectiveness study are used to illustrate key concepts including: communication of model specification, parameter interpretation, sample size and power, and missing data. Input and key output files from Mplus and SAS are provided. A cluster randomized trial with repeated measures (i.e., three-level regression model) is then briefly presented with simulated data based on some core ideas from a cognitive behavioral stress management intervention in prostate cancer. PMID:23107843

Using mixed treatment comparisons and meta-regression to perform indirect comparisons to estimate the efficacy of biologic treatments in rheumatoid arthritis.

PubMed

Nixon, R M; Bansback, N; Brennan, A

2007-03-15

Mixed treatment comparison (MTC) is a generalization of meta-analysis. Instead of the same treatment for a disease being tested in a number of studies, a number of different interventions are considered. Meta-regression is also a generalization of meta-analysis where an attempt is made to explain the heterogeneity between the treatment effects in the studies by regressing on study-level covariables. Our focus is where there are several different treatments considered in a number of randomized controlled trials in a specific disease, the same treatment can be applied in several arms within a study, and where differences in efficacy can be explained by differences in the study settings. We develop methods for simultaneously comparing several treatments and adjusting for study-level covariables by combining ideas from MTC and meta-regression. We use a case study from rheumatoid arthritis. We identified relevant trials of biologic verses standard therapy or placebo and extracted the doses, comparators and patient baseline characteristics. Efficacy is measured using the log odds ratio of achieving six-month ACR50 responder status. A random-effects meta-regression model is fitted which adjusts the log odds ratio for study-level prognostic factors. A different random-effect distribution on the log odds ratios is allowed for each different treatment. The odds ratio is found as a function of the prognostic factors for each treatment. The apparent differences in the randomized trials between tumour necrosis factor alpha (TNF- alpha) antagonists are explained by differences in prognostic factors and the analysis suggests that these drugs as a class are not different from each other. Copyright (c) 2006 John Wiley & Sons, Ltd.

Regression-assisted deconvolution.

PubMed

McIntyre, Julie; Stefanski, Leonard A

2011-06-30

We present a semi-parametric deconvolution estimator for the density function of a random variable biX that is measured with error, a common challenge in many epidemiological studies. Traditional deconvolution estimators rely only on assumptions about the distribution of X and the error in its measurement, and ignore information available in auxiliary variables. Our method assumes the availability of a covariate vector statistically related to X by a mean-variance function regression model, where regression errors are normally distributed and independent of the measurement errors. Simulations suggest that the estimator achieves a much lower integrated squared error than the observed-data kernel density estimator when models are correctly specified and the assumption of normal regression errors is met. We illustrate the method using anthropometric measurements of newborns to estimate the density function of newborn length. Copyright © 2011 John Wiley & Sons, Ltd.

DOE Office of Scientific and Technical Information (OSTI.GOV)

Tang, Kunkun, E-mail: ktg@illinois.edu; Inria Bordeaux – Sud-Ouest, Team Cardamom, 200 avenue de la Vieille Tour, 33405 Talence; Congedo, Pietro M.

The Polynomial Dimensional Decomposition (PDD) is employed in this work for the global sensitivity analysis and uncertainty quantification (UQ) of stochastic systems subject to a moderate to large number of input random variables. Due to the intimate connection between the PDD and the Analysis of Variance (ANOVA) approaches, PDD is able to provide a simpler and more direct evaluation of the Sobol' sensitivity indices, when compared to the Polynomial Chaos expansion (PC). Unfortunately, the number of PDD terms grows exponentially with respect to the size of the input random vector, which makes the computational cost of standard methods unaffordable formore » real engineering applications. In order to address the problem of the curse of dimensionality, this work proposes essentially variance-based adaptive strategies aiming to build a cheap meta-model (i.e. surrogate model) by employing the sparse PDD approach with its coefficients computed by regression. Three levels of adaptivity are carried out in this paper: 1) the truncated dimensionality for ANOVA component functions, 2) the active dimension technique especially for second- and higher-order parameter interactions, and 3) the stepwise regression approach designed to retain only the most influential polynomials in the PDD expansion. During this adaptive procedure featuring stepwise regressions, the surrogate model representation keeps containing few terms, so that the cost to resolve repeatedly the linear systems of the least-squares regression problem is negligible. The size of the finally obtained sparse PDD representation is much smaller than the one of the full expansion, since only significant terms are eventually retained. Consequently, a much smaller number of calls to the deterministic model is required to compute the final PDD coefficients.« less

Design of Probabilistic Random Forests with Applications to Anticancer Drug Sensitivity Prediction

PubMed Central

Rahman, Raziur; Haider, Saad; Ghosh, Souparno; Pal, Ranadip

2015-01-01

Random forests consisting of an ensemble of regression trees with equal weights are frequently used for design of predictive models. In this article, we consider an extension of the methodology by representing the regression trees in the form of probabilistic trees and analyzing the nature of heteroscedasticity. The probabilistic tree representation allows for analytical computation of confidence intervals (CIs), and the tree weight optimization is expected to provide stricter CIs with comparable performance in mean error. We approached the ensemble of probabilistic trees’ prediction from the perspectives of a mixture distribution and as a weighted sum of correlated random variables. We applied our methodology to the drug sensitivity prediction problem on synthetic and cancer cell line encyclopedia dataset and illustrated that tree weights can be selected to reduce the average length of the CI without increase in mean error. PMID:27081304

Quality Quandaries: Predicting a Population of Curves

DOE PAGES

Fugate, Michael Lynn; Hamada, Michael Scott; Weaver, Brian Phillip

2017-12-19

We present a random effects spline regression model based on splines that provides an integrated approach for analyzing functional data, i.e., curves, when the shape of the curves is not parametrically specified. An analysis using this model is presented that makes inferences about a population of curves as well as features of the curves.

Quality Quandaries: Predicting a Population of Curves

DOE Office of Scientific and Technical Information (OSTI.GOV)

Fugate, Michael Lynn; Hamada, Michael Scott; Weaver, Brian Phillip

We present a random effects spline regression model based on splines that provides an integrated approach for analyzing functional data, i.e., curves, when the shape of the curves is not parametrically specified. An analysis using this model is presented that makes inferences about a population of curves as well as features of the curves.

Assessing Mediation Using Marginal Structural Models in the Presence of Confounding and Moderation

ERIC Educational Resources Information Center

Coffman, Donna L.; Zhong, Wei

2012-01-01

This article presents marginal structural models with inverse propensity weighting (IPW) for assessing mediation. Generally, individuals are not randomly assigned to levels of the mediator. Therefore, confounders of the mediator and outcome may exist that limit causal inferences, a goal of mediation analysis. Either regression adjustment or IPW…

Multitrait, random regression, or simple repeatability model in high-throughput phenotyping data improve genomic prediction for wheat grain yield

USDA-ARS?s Scientific Manuscript database

High-throughput phenotyping (HTP) platforms can be used to measure traits that are genetically correlated with wheat (Triticum aestivum L.) grain yield across time. Incorporating such secondary traits in the multivariate pedigree and genomic prediction models would be desirable to improve indirect s...

Stemflow estimation in a redwood forest using model-based stratified random sampling

Treesearch

Jack Lewis

2003-01-01

Model-based stratified sampling is illustrated by a case study of stemflow volume in a redwood forest. The approach is actually a model-assisted sampling design in which auxiliary information (tree diameter) is utilized in the design of stratum boundaries to optimize the efficiency of a regression or ratio estimator. The auxiliary information is utilized in both the...

Considerations for analysis of time-to-event outcomes measured with error: Bias and correction with SIMEX.

PubMed

Oh, Eric J; Shepherd, Bryan E; Lumley, Thomas; Shaw, Pamela A

2018-04-15

For time-to-event outcomes, a rich literature exists on the bias introduced by covariate measurement error in regression models, such as the Cox model, and methods of analysis to address this bias. By comparison, less attention has been given to understanding the impact or addressing errors in the failure time outcome. For many diseases, the timing of an event of interest (such as progression-free survival or time to AIDS progression) can be difficult to assess or reliant on self-report and therefore prone to measurement error. For linear models, it is well known that random errors in the outcome variable do not bias regression estimates. With nonlinear models, however, even random error or misclassification can introduce bias into estimated parameters. We compare the performance of 2 common regression models, the Cox and Weibull models, in the setting of measurement error in the failure time outcome. We introduce an extension of the SIMEX method to correct for bias in hazard ratio estimates from the Cox model and discuss other analysis options to address measurement error in the response. A formula to estimate the bias induced into the hazard ratio by classical measurement error in the event time for a log-linear survival model is presented. Detailed numerical studies are presented to examine the performance of the proposed SIMEX method under varying levels and parametric forms of the error in the outcome. We further illustrate the method with observational data on HIV outcomes from the Vanderbilt Comprehensive Care Clinic. Copyright © 2017 John Wiley & Sons, Ltd.

Sentinel node status prediction by four statistical models: results from a large bi-institutional series (n = 1132).

PubMed

Mocellin, Simone; Thompson, John F; Pasquali, Sandro; Montesco, Maria C; Pilati, Pierluigi; Nitti, Donato; Saw, Robyn P; Scolyer, Richard A; Stretch, Jonathan R; Rossi, Carlo R

2009-12-01

To improve selection for sentinel node (SN) biopsy (SNB) in patients with cutaneous melanoma using statistical models predicting SN status. About 80% of patients currently undergoing SNB are node negative. In the absence of conclusive evidence of a SNBassociated survival benefit, these patients may be over-treated. Here, we tested the efficiency of 4 different models in predicting SN status. The clinicopathologic data (age, gender, tumor thickness, Clark level, regression, ulceration, histologic subtype, and mitotic index) of 1132 melanoma patients who had undergone SNB at institutions in Italy and Australia were analyzed. Logistic regression, classification tree, random forest, and support vector machine models were fitted to the data. The predictive models were built with the aim of maximizing the negative predictive value (NPV) and reducing the rate of SNB procedures though minimizing the error rate. After cross-validation logistic regression, classification tree, random forest, and support vector machine predictive models obtained clinically relevant NPV (93.6%, 94.0%, 97.1%, and 93.0%, respectively), SNB reduction (27.5%, 29.8%, 18.2%, and 30.1%, respectively), and error rates (1.8%, 1.8%, 0.5%, and 2.1%, respectively). Using commonly available clinicopathologic variables, predictive models can preoperatively identify a proportion of patients ( approximately 25%) who might be spared SNB, with an acceptable (1%-2%) error. If validated in large prospective series, these models might be implemented in the clinical setting for improved patient selection, which ultimately would lead to better quality of life for patients and optimization of resource allocation for the health care system.

Adaptive surrogate modeling by ANOVA and sparse polynomial dimensional decomposition for global sensitivity analysis in fluid simulation

NASA Astrophysics Data System (ADS)

Tang, Kunkun; Congedo, Pietro M.; Abgrall, Rémi

2016-06-01

The Polynomial Dimensional Decomposition (PDD) is employed in this work for the global sensitivity analysis and uncertainty quantification (UQ) of stochastic systems subject to a moderate to large number of input random variables. Due to the intimate connection between the PDD and the Analysis of Variance (ANOVA) approaches, PDD is able to provide a simpler and more direct evaluation of the Sobol' sensitivity indices, when compared to the Polynomial Chaos expansion (PC). Unfortunately, the number of PDD terms grows exponentially with respect to the size of the input random vector, which makes the computational cost of standard methods unaffordable for real engineering applications. In order to address the problem of the curse of dimensionality, this work proposes essentially variance-based adaptive strategies aiming to build a cheap meta-model (i.e. surrogate model) by employing the sparse PDD approach with its coefficients computed by regression. Three levels of adaptivity are carried out in this paper: 1) the truncated dimensionality for ANOVA component functions, 2) the active dimension technique especially for second- and higher-order parameter interactions, and 3) the stepwise regression approach designed to retain only the most influential polynomials in the PDD expansion. During this adaptive procedure featuring stepwise regressions, the surrogate model representation keeps containing few terms, so that the cost to resolve repeatedly the linear systems of the least-squares regression problem is negligible. The size of the finally obtained sparse PDD representation is much smaller than the one of the full expansion, since only significant terms are eventually retained. Consequently, a much smaller number of calls to the deterministic model is required to compute the final PDD coefficients.

Random forest regression modelling for forest aboveground biomass estimation using RISAT-1 PolSAR and terrestrial LiDAR data

NASA Astrophysics Data System (ADS)

Mangla, Rohit; Kumar, Shashi; Nandy, Subrata

2016-05-01

SAR and LiDAR remote sensing have already shown the potential of active sensors for forest parameter retrieval. SAR sensor in its fully polarimetric mode has an advantage to retrieve scattering property of different component of forest structure and LiDAR has the capability to measure structural information with very high accuracy. This study was focused on retrieval of forest aboveground biomass (AGB) using Terrestrial Laser Scanner (TLS) based point clouds and scattering property of forest vegetation obtained from decomposition modelling of RISAT-1 fully polarimetric SAR data. TLS data was acquired for 14 plots of Timli forest range, Uttarakhand, India. The forest area is dominated by Sal trees and random sampling with plot size of 0.1 ha (31.62m*31.62m) was adopted for TLS and field data collection. RISAT-1 data was processed to retrieve SAR data based variables and TLS point clouds based 3D imaging was done to retrieve LiDAR based variables. Surface scattering, double-bounce scattering, volume scattering, helix and wire scattering were the SAR based variables retrieved from polarimetric decomposition. Tree heights and stem diameters were used as LiDAR based variables retrieved from single tree vertical height and least square circle fit methods respectively. All the variables obtained for forest plots were used as an input in a machine learning based Random Forest Regression Model, which was developed in this study for forest AGB estimation. Modelled output for forest AGB showed reliable accuracy (RMSE = 27.68 t/ha) and a good coefficient of determination (0.63) was obtained through the linear regression between modelled AGB and field-estimated AGB. The sensitivity analysis showed that the model was more sensitive for the major contributed variables (stem diameter and volume scattering) and these variables were measured from two different remote sensing techniques. This study strongly recommends the integration of SAR and LiDAR data for forest AGB estimation.

Why choose Random Forest to predict rare species distribution with few samples in large undersampled areas? Three Asian crane species models provide supporting evidence.

PubMed

Mi, Chunrong; Huettmann, Falk; Guo, Yumin; Han, Xuesong; Wen, Lijia

2017-01-01

Species distribution models (SDMs) have become an essential tool in ecology, biogeography, evolution and, more recently, in conservation biology. How to generalize species distributions in large undersampled areas, especially with few samples, is a fundamental issue of SDMs. In order to explore this issue, we used the best available presence records for the Hooded Crane ( Grus monacha , n  = 33), White-naped Crane ( Grus vipio , n  = 40), and Black-necked Crane ( Grus nigricollis , n  = 75) in China as three case studies, employing four powerful and commonly used machine learning algorithms to map the breeding distributions of the three species: TreeNet (Stochastic Gradient Boosting, Boosted Regression Tree Model), Random Forest, CART (Classification and Regression Tree) and Maxent (Maximum Entropy Models). In addition, we developed an ensemble forecast by averaging predicted probability of the above four models results. Commonly used model performance metrics (Area under ROC (AUC) and true skill statistic (TSS)) were employed to evaluate model accuracy. The latest satellite tracking data and compiled literature data were used as two independent testing datasets to confront model predictions. We found Random Forest demonstrated the best performance for the most assessment method, provided a better model fit to the testing data, and achieved better species range maps for each crane species in undersampled areas. Random Forest has been generally available for more than 20 years and has been known to perform extremely well in ecological predictions. However, while increasingly on the rise, its potential is still widely underused in conservation, (spatial) ecological applications and for inference. Our results show that it informs ecological and biogeographical theories as well as being suitable for conservation applications, specifically when the study area is undersampled. This method helps to save model-selection time and effort, and allows robust and rapid assessments and decisions for efficient conservation.

Why choose Random Forest to predict rare species distribution with few samples in large undersampled areas? Three Asian crane species models provide supporting evidence

PubMed Central

Mi, Chunrong; Huettmann, Falk; Han, Xuesong; Wen, Lijia

2017-01-01

Species distribution models (SDMs) have become an essential tool in ecology, biogeography, evolution and, more recently, in conservation biology. How to generalize species distributions in large undersampled areas, especially with few samples, is a fundamental issue of SDMs. In order to explore this issue, we used the best available presence records for the Hooded Crane (Grus monacha, n = 33), White-naped Crane (Grus vipio, n = 40), and Black-necked Crane (Grus nigricollis, n = 75) in China as three case studies, employing four powerful and commonly used machine learning algorithms to map the breeding distributions of the three species: TreeNet (Stochastic Gradient Boosting, Boosted Regression Tree Model), Random Forest, CART (Classification and Regression Tree) and Maxent (Maximum Entropy Models). In addition, we developed an ensemble forecast by averaging predicted probability of the above four models results. Commonly used model performance metrics (Area under ROC (AUC) and true skill statistic (TSS)) were employed to evaluate model accuracy. The latest satellite tracking data and compiled literature data were used as two independent testing datasets to confront model predictions. We found Random Forest demonstrated the best performance for the most assessment method, provided a better model fit to the testing data, and achieved better species range maps for each crane species in undersampled areas. Random Forest has been generally available for more than 20 years and has been known to perform extremely well in ecological predictions. However, while increasingly on the rise, its potential is still widely underused in conservation, (spatial) ecological applications and for inference. Our results show that it informs ecological and biogeographical theories as well as being suitable for conservation applications, specifically when the study area is undersampled. This method helps to save model-selection time and effort, and allows robust and rapid assessments and decisions for efficient conservation. PMID:28097060

FOG Random Drift Signal Denoising Based on the Improved AR Model and Modified Sage-Husa Adaptive Kalman Filter.

PubMed

Sun, Jin; Xu, Xiaosu; Liu, Yiting; Zhang, Tao; Li, Yao

2016-07-12

In order to reduce the influence of fiber optic gyroscope (FOG) random drift error on inertial navigation systems, an improved auto regressive (AR) model is put forward in this paper. First, based on real-time observations at each restart of the gyroscope, the model of FOG random drift can be established online. In the improved AR model, the FOG measured signal is employed instead of the zero mean signals. Then, the modified Sage-Husa adaptive Kalman filter (SHAKF) is introduced, which can directly carry out real-time filtering on the FOG signals. Finally, static and dynamic experiments are done to verify the effectiveness. The filtering results are analyzed with Allan variance. The analysis results show that the improved AR model has high fitting accuracy and strong adaptability, and the minimum fitting accuracy of single noise is 93.2%. Based on the improved AR(3) model, the denoising method of SHAKF is more effective than traditional methods, and its effect is better than 30%. The random drift error of FOG is reduced effectively, and the precision of the FOG is improved.

Use of AMMI and linear regression models to analyze genotype-environment interaction in durum wheat.

PubMed

Nachit, M M; Nachit, G; Ketata, H; Gauch, H G; Zobel, R W

1992-03-01

The joint durum wheat (Triticum turgidum L var 'durum') breeding program of the International Maize and Wheat Improvement Center (CIMMYT) and the International Center for Agricultural Research in the Dry Areas (ICARDA) for the Mediterranean region employs extensive multilocation testing. Multilocation testing produces significant genotype-environment (GE) interaction that reduces the accuracy for estimating yield and selecting appropriate germ plasm. The sum of squares (SS) of GE interaction was partitioned by linear regression techniques into joint, genotypic, and environmental regressions, and by Additive Main effects and the Multiplicative Interactions (AMMI) model into five significant Interaction Principal Component Axes (IPCA). The AMMI model was more effective in partitioning the interaction SS than the linear regression technique. The SS contained in the AMMI model was 6 times higher than the SS for all three regressions. Postdictive assessment recommended the use of the first five IPCA axes, while predictive assessment AMMI1 (main effects plus IPCA1). After elimination of random variation, AMMI1 estimates for genotypic yields within sites were more precise than unadjusted means. This increased precision was equivalent to increasing the number of replications by a factor of 3.7.

A nonparametric multiple imputation approach for missing categorical data.

PubMed

Zhou, Muhan; He, Yulei; Yu, Mandi; Hsu, Chiu-Hsieh

2017-06-06

Incomplete categorical variables with more than two categories are common in public health data. However, most of the existing missing-data methods do not use the information from nonresponse (missingness) probabilities. We propose a nearest-neighbour multiple imputation approach to impute a missing at random categorical outcome and to estimate the proportion of each category. The donor set for imputation is formed by measuring distances between each missing value with other non-missing values. The distance function is calculated based on a predictive score, which is derived from two working models: one fits a multinomial logistic regression for predicting the missing categorical outcome (the outcome model) and the other fits a logistic regression for predicting missingness probabilities (the missingness model). A weighting scheme is used to accommodate contributions from two working models when generating the predictive score. A missing value is imputed by randomly selecting one of the non-missing values with the smallest distances. We conduct a simulation to evaluate the performance of the proposed method and compare it with several alternative methods. A real-data application is also presented. The simulation study suggests that the proposed method performs well when missingness probabilities are not extreme under some misspecifications of the working models. However, the calibration estimator, which is also based on two working models, can be highly unstable when missingness probabilities for some observations are extremely high. In this scenario, the proposed method produces more stable and better estimates. In addition, proper weights need to be chosen to balance the contributions from the two working models and achieve optimal results for the proposed method. We conclude that the proposed multiple imputation method is a reasonable approach to dealing with missing categorical outcome data with more than two levels for assessing the distribution of the outcome. In terms of the choices for the working models, we suggest a multinomial logistic regression for predicting the missing outcome and a binary logistic regression for predicting the missingness probability.

Analytical framework for reconstructing heterogeneous environmental variables from mammal community structure.

PubMed

Louys, Julien; Meloro, Carlo; Elton, Sarah; Ditchfield, Peter; Bishop, Laura C

2015-01-01

We test the performance of two models that use mammalian communities to reconstruct multivariate palaeoenvironments. While both models exploit the correlation between mammal communities (defined in terms of functional groups) and arboreal heterogeneity, the first uses a multiple multivariate regression of community structure and arboreal heterogeneity, while the second uses a linear regression of the principal components of each ecospace. The success of these methods means the palaeoenvironment of a particular locality can be reconstructed in terms of the proportions of heavy, moderate, light, and absent tree canopy cover. The linear regression is less biased, and more precisely and accurately reconstructs heavy tree canopy cover than the multiple multivariate model. However, the multiple multivariate model performs better than the linear regression for all other canopy cover categories. Both models consistently perform better than randomly generated reconstructions. We apply both models to the palaeocommunity of the Upper Laetolil Beds, Tanzania. Our reconstructions indicate that there was very little heavy tree cover at this site (likely less than 10%), with the palaeo-landscape instead comprising a mixture of light and absent tree cover. These reconstructions help resolve the previous conflicting palaeoecological reconstructions made for this site. Copyright © 2014 Elsevier Ltd. All rights reserved.

Personality predicts time to remission and clinical status in hypochondriasis during a 6-year follow-up.

PubMed

Greeven, Anja; van Balkom, Anton J L M; Spinhoven, Philip

2014-05-01

We aimed to investigate whether personality characteristics predict time to remission and psychiatric status. The follow-up was at most 6 years and was performed within the scope of a randomized controlled trial that investigated the efficacy of cognitive behavioral therapy, paroxetine, and placebo in hypochondriasis. The Life Chart Interview was administered to investigate for each year if remission had occurred. Personality was assessed at pretest by the Abbreviated Dutch Temperament and Character Inventory. Cox's regression models for recurrent events were compared with logistic regression models. Sixteen (36.4%) of 44 patients achieved remission during the follow-up period. Cox's regression yielded approximately the same results as the logistic regression. Being less harm avoidant and more cooperative were associated with a shorter time to remission and a remitted state after the follow-up period. Personality variables seem to be relevant for describing patients with a more chronic course of hypochondriacal complaints.

Input variable selection and calibration data selection for storm water quality regression models.

PubMed

Sun, Siao; Bertrand-Krajewski, Jean-Luc

2013-01-01

Storm water quality models are useful tools in storm water management. Interest has been growing in analyzing existing data for developing models for urban storm water quality evaluations. It is important to select appropriate model inputs when many candidate explanatory variables are available. Model calibration and verification are essential steps in any storm water quality modeling. This study investigates input variable selection and calibration data selection in storm water quality regression models. The two selection problems are mutually interacted. A procedure is developed in order to fulfil the two selection tasks in order. The procedure firstly selects model input variables using a cross validation method. An appropriate number of variables are identified as model inputs to ensure that a model is neither overfitted nor underfitted. Based on the model input selection results, calibration data selection is studied. Uncertainty of model performances due to calibration data selection is investigated with a random selection method. An approach using the cluster method is applied in order to enhance model calibration practice based on the principle of selecting representative data for calibration. The comparison between results from the cluster selection method and random selection shows that the former can significantly improve performances of calibrated models. It is found that the information content in calibration data is important in addition to the size of calibration data.

Comparison of methods for the analysis of relatively simple mediation models.

PubMed

Rijnhart, Judith J M; Twisk, Jos W R; Chinapaw, Mai J M; de Boer, Michiel R; Heymans, Martijn W

2017-09-01

Statistical mediation analysis is an often used method in trials, to unravel the pathways underlying the effect of an intervention on a particular outcome variable. Throughout the years, several methods have been proposed, such as ordinary least square (OLS) regression, structural equation modeling (SEM), and the potential outcomes framework. Most applied researchers do not know that these methods are mathematically equivalent when applied to mediation models with a continuous mediator and outcome variable. Therefore, the aim of this paper was to demonstrate the similarities between OLS regression, SEM, and the potential outcomes framework in three mediation models: 1) a crude model, 2) a confounder-adjusted model, and 3) a model with an interaction term for exposure-mediator interaction. Secondary data analysis of a randomized controlled trial that included 546 schoolchildren. In our data example, the mediator and outcome variable were both continuous. We compared the estimates of the total, direct and indirect effects, proportion mediated, and 95% confidence intervals (CIs) for the indirect effect across OLS regression, SEM, and the potential outcomes framework. OLS regression, SEM, and the potential outcomes framework yielded the same effect estimates in the crude mediation model, the confounder-adjusted mediation model, and the mediation model with an interaction term for exposure-mediator interaction. Since OLS regression, SEM, and the potential outcomes framework yield the same results in three mediation models with a continuous mediator and outcome variable, researchers can continue using the method that is most convenient to them.

Prediction of Return-to-original-work after an Industrial Accident Using Machine Learning and Comparison of Techniques

PubMed Central

2018-01-01

Background Many studies have tried to develop predictors for return-to-work (RTW). However, since complex factors have been demonstrated to predict RTW, it is difficult to use them practically. This study investigated whether factors used in previous studies could predict whether an individual had returned to his/her original work by four years after termination of the worker's recovery period. Methods An initial logistic regression analysis of 1,567 participants of the fourth Panel Study of Worker's Compensation Insurance yielded odds ratios. The participants were divided into two subsets, a training dataset and a test dataset. Using the training dataset, logistic regression, decision tree, random forest, and support vector machine models were established, and important variables of each model were identified. The predictive abilities of the different models were compared. Results The analysis showed that only earned income and company-related factors significantly affected return-to-original-work (RTOW). The random forest model showed the best accuracy among the tested machine learning models; however, the difference was not prominent. Conclusion It is possible to predict a worker's probability of RTOW using machine learning techniques with moderate accuracy. PMID:29736160

[Prediction model of health workforce and beds in county hospitals of Hunan by multiple linear regression].

PubMed

Ling, Ru; Liu, Jiawang

2011-12-01

To construct prediction model for health workforce and hospital beds in county hospitals of Hunan by multiple linear regression. We surveyed 16 counties in Hunan with stratified random sampling according to uniform questionnaires,and multiple linear regression analysis with 20 quotas selected by literature view was done. Independent variables in the multiple linear regression model on medical personnels in county hospitals included the counties' urban residents' income, crude death rate, medical beds, business occupancy, professional equipment value, the number of devices valued above 10 000 yuan, fixed assets, long-term debt, medical income, medical expenses, outpatient and emergency visits, hospital visits, actual available bed days, and utilization rate of hospital beds. Independent variables in the multiple linear regression model on county hospital beds included the the population of aged 65 and above in the counties, disposable income of urban residents, medical personnel of medical institutions in county area, business occupancy, the total value of professional equipment, fixed assets, long-term debt, medical income, medical expenses, outpatient and emergency visits, hospital visits, actual available bed days, utilization rate of hospital beds, and length of hospitalization. The prediction model shows good explanatory and fitting, and may be used for short- and mid-term forecasting.

Modeling species’ realized climatic niche space and predicting their response to global warming for several western forest species with small geographic distributions

Treesearch

Marcus V. Warwell; Gerald E. Rehfeldt; Nicholas L. Crookston

2010-01-01

The Random Forests multiple regression tree was used to develop an empirically based bioclimatic model of the presence-absence of species occupying small geographic distributions in western North America. The species assessed were subalpine larch (Larix lyallii), smooth Arizona cypress (Cupressus arizonica ssp. glabra...

A Nationwide Epidemiologic Modeling Study of LD: Risk, Protection, and Unintended Impact

ERIC Educational Resources Information Center

McDermott, Paul A.; Goldberg, Michelle M.; Watkins, Marley W.; Stanley, Jeanne L.; Glutting, Joseph J.

2006-01-01

Through multiple logistic regression modeling, this article explores the relative importance of risk and protective factors associated with learning disabilities (LD). A representative national sample of 6- to 17-year-old students (N = 1,268) was drawn by random stratification and classified by the presence versus absence of LD in reading,…

Black Clouds vs Random Variation in Hospital Admissions.

PubMed

Ong, Luei Wern; Dawson, Jeffrey D; Ely, John W

2018-06-01

Physicians often accuse their peers of being "black clouds" if they repeatedly have more than the average number of hospital admissions while on call. Our purpose was to determine whether the black-cloud phenomenon is real or explainable by random variation. We analyzed hospital admissions to the University of Iowa family medicine service from July 1, 2010 to June 30, 2015. Analyses were stratified by peer group (eg, night shift attending physicians, day shift senior residents). We analyzed admission numbers to find evidence of black-cloud physicians (those with significantly more admissions than their peers) and white-cloud physicians (those with significantly fewer admissions). The statistical significance of whether there were actual differences across physicians was tested with mixed-effects negative binomial regression. The 5-year study included 96 physicians and 6,194 admissions. The number of daytime admissions ranged from 0 to 10 (mean 2.17, SD 1.63). Night admissions ranged from 0 to 11 (mean 1.23, SD 1.22). Admissions increased from 1,016 in the first year to 1,523 in the fifth year. We found 18 white-cloud and 16 black-cloud physicians in simple regression models that did not control for this upward trend. After including study year and other potential confounding variables in the regression models, there were no significant associations between physicians and admission numbers and therefore no true black or white clouds. In this study, apparent black-cloud and white-cloud physicians could be explained by random variation in hospital admissions. However, this randomness incorporated a wide range in workload among physicians, with potential impact on resident education at the low end and patient safety at the high end.

Modeling Verdict Outcomes Using Social Network Measures: The Watergate and Caviar Network Cases

PubMed Central

2016-01-01

Modelling criminal trial verdict outcomes using social network measures is an emerging research area in quantitative criminology. Few studies have yet analyzed which of these measures are the most important for verdict modelling or which data classification techniques perform best for this application. To compare the performance of different techniques in classifying members of a criminal network, this article applies three different machine learning classifiers–Logistic Regression, Naïve Bayes and Random Forest–with a range of social network measures and the necessary databases to model the verdicts in two real–world cases: the U.S. Watergate Conspiracy of the 1970’s and the now–defunct Canada–based international drug trafficking ring known as the Caviar Network. In both cases it was found that the Random Forest classifier did better than either Logistic Regression or Naïve Bayes, and its superior performance was statistically significant. This being so, Random Forest was used not only for classification but also to assess the importance of the measures. For the Watergate case, the most important one proved to be betweenness centrality while for the Caviar Network, it was the effective size of the network. These results are significant because they show that an approach combining machine learning with social network analysis not only can generate accurate classification models but also helps quantify the importance social network variables in modelling verdict outcomes. We conclude our analysis with a discussion and some suggestions for future work in verdict modelling using social network measures. PMID:26824351

Modeling landslide susceptibility in data-scarce environments using optimized data mining and statistical methods

NASA Astrophysics Data System (ADS)

Lee, Jung-Hyun; Sameen, Maher Ibrahim; Pradhan, Biswajeet; Park, Hyuck-Jin

2018-02-01

This study evaluated the generalizability of five models to select a suitable approach for landslide susceptibility modeling in data-scarce environments. In total, 418 landslide inventories and 18 landslide conditioning factors were analyzed. Multicollinearity and factor optimization were investigated before data modeling, and two experiments were then conducted. In each experiment, five susceptibility maps were produced based on support vector machine (SVM), random forest (RF), weight-of-evidence (WoE), ridge regression (Rid_R), and robust regression (RR) models. The highest accuracy (AUC = 0.85) was achieved with the SVM model when either the full or limited landslide inventories were used. Furthermore, the RF and WoE models were severely affected when less landslide samples were used for training. The other models were affected slightly when the training samples were limited.

The Past, Present and Future of the Meteorological Phenomena Identification Near the Ground (mPING) Project

NASA Astrophysics Data System (ADS)

Elmore, K. L.

2016-12-01

The Metorological Phenomemna Identification NeartheGround (mPING) project is an example of a crowd-sourced, citizen science effort to gather data of sufficeint quality and quantity needed by new post processing methods that use machine learning. Transportation and infrastructure are particularly sensitive to precipitation type in winter weather. We extract attributes from operational numerical forecast models and use them in a random forest to generate forecast winter precipitation types. We find that random forests applied to forecast soundings are effective at generating skillful forecasts of surface ptype with consideralbly more skill than the current algorithms, especuially for ice pellets and freezing rain. We also find that three very different forecast models yuield similar overall results, showing that random forests are able to extract essentially equivalent information from different forecast models. We also show that the random forest for each model, and each profile type is unique to the particular forecast model and that the random forests developed using a particular model suffer significant degradation when given attributes derived from a different model. This implies that no single algorithm can perform well across all forecast models. Clearly, random forests extract information unavailable to "physically based" methods because the physical information in the models does not appear as we expect. One intersting result is that results from the classic "warm nose" sounding profile are, by far, the most sensitive to the particular forecast model, but this profile is also the one for which random forests are most skillful. Finally, a method for calibrarting probabilties for each different ptype using multinomial logistic regression is shown.

Genetic evaluation and selection response for growth in meat-type quail through random regression models using B-spline functions and Legendre polynomials.

PubMed

Mota, L F M; Martins, P G M A; Littiere, T O; Abreu, L R A; Silva, M A; Bonafé, C M

2018-04-01

The objective was to estimate (co)variance functions using random regression models (RRM) with Legendre polynomials, B-spline function and multi-trait models aimed at evaluating genetic parameters of growth traits in meat-type quail. A database containing the complete pedigree information of 7000 meat-type quail was utilized. The models included the fixed effects of contemporary group and generation. Direct additive genetic and permanent environmental effects, considered as random, were modeled using B-spline functions considering quadratic and cubic polynomials for each individual segment, and Legendre polynomials for age. Residual variances were grouped in four age classes. Direct additive genetic and permanent environmental effects were modeled using 2 to 4 segments and were modeled by Legendre polynomial with orders of fit ranging from 2 to 4. The model with quadratic B-spline adjustment, using four segments for direct additive genetic and permanent environmental effects, was the most appropriate and parsimonious to describe the covariance structure of the data. The RRM using Legendre polynomials presented an underestimation of the residual variance. Lesser heritability estimates were observed for multi-trait models in comparison with RRM for the evaluated ages. In general, the genetic correlations between measures of BW from hatching to 35 days of age decreased as the range between the evaluated ages increased. Genetic trend for BW was positive and significant along the selection generations. The genetic response to selection for BW in the evaluated ages presented greater values for RRM compared with multi-trait models. In summary, RRM using B-spline functions with four residual variance classes and segments were the best fit for genetic evaluation of growth traits in meat-type quail. In conclusion, RRM should be considered in genetic evaluation of breeding programs.

[Associations between dormitory environment/other factors and sleep quality of medical students].

PubMed

Zheng, Bang; Wang, Kailu; Pan, Ziqi; Li, Man; Pan, Yuting; Liu, Ting; Xu, Dan; Lyu, Jun

2016-03-01

To investigate the sleep quality and related factors among medical students in China, understand the association between dormitory environment and sleep quality, and provide evidence and recommendations for sleep hygiene intervention. A total of 555 undergraduate students were selected from a medical school of an university in Beijing through stratified-cluster random-sampling to conduct a questionnaire survey by using Chinese version of Pittsburgh Sleep Quality Index (PSQI) and self-designed questionnaire. Analyses were performed by using multiple logistic regression model as well as multilevel linear regression model. The prevalence of sleep disorder was 29.1%(149/512), and 39.1%(200/512) of the students reported that the sleep quality was influenced by dormitory environment. PSQI score was negatively correlated with self-reported rating of dormitory environment (γs=-0.310, P<0.001). Logistic regression analysis showed the related factors of sleep disorder included grade, sleep regularity, self-rated health status, pressures of school work and employment, as well as dormitory environment. RESULTS of multilevel regression analysis also indicated that perception on dormitory environment (individual level) was associated with sleep quality with the dormitory level random effects under control (b=-0.619, P<0.001). The prevalence of sleep disorder was high in medical students, which was associated with multiple factors. Dormitory environment should be taken into consideration when the interventions are taken to improve the sleep quality of students.

Genetic background in partitioning of metabolizable energy efficiency in dairy cows.

PubMed

Mehtiö, T; Negussie, E; Mäntysaari, P; Mäntysaari, E A; Lidauer, M H

2018-05-01

The main objective of this study was to assess the genetic differences in metabolizable energy efficiency and efficiency in partitioning metabolizable energy in different pathways: maintenance, milk production, and growth in primiparous dairy cows. Repeatability models for residual energy intake (REI) and metabolizable energy intake (MEI) were compared and the genetic and permanent environmental variations in MEI were partitioned into its energy sinks using random regression models. We proposed 2 new feed efficiency traits: metabolizable energy efficiency (MEE), which is formed by modeling MEI fitting regressions on energy sinks [metabolic body weight (BW 0.75 ), energy-corrected milk, body weight gain, and body weight loss] directly; and partial MEE (pMEE), where the model for MEE is extended with regressions on energy sinks nested within additive genetic and permanent environmental effects. The data used were collected from Luke's experimental farms Rehtijärvi and Minkiö between 1998 and 2014. There were altogether 12,350 weekly MEI records on 495 primiparous Nordic Red dairy cows from wk 2 to 40 of lactation. Heritability estimates for REI and MEE were moderate, 0.33 and 0.26, respectively. The estimate of the residual variance was smaller for MEE than for REI, indicating that analyzing weekly MEI observations simultaneously with energy sinks is preferable. Model validation based on Akaike's information criterion showed that pMEE models fitted the data even better and also resulted in smaller residual variance estimates. However, models that included random regression on BW 0.75 converged slowly. The resulting genetic standard deviation estimate from the pMEE coefficient for milk production was 0.75 MJ of MEI/kg of energy-corrected milk. The derived partial heritabilities for energy efficiency in maintenance, milk production, and growth were 0.02, 0.06, and 0.04, respectively, indicating that some genetic variation may exist in the efficiency of using metabolizable energy for different pathways in dairy cows. Copyright © 2018 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

A menu-driven software package of Bayesian nonparametric (and parametric) mixed models for regression analysis and density estimation.

PubMed

Karabatsos, George

2017-02-01

Most of applied statistics involves regression analysis of data. In practice, it is important to specify a regression model that has minimal assumptions which are not violated by data, to ensure that statistical inferences from the model are informative and not misleading. This paper presents a stand-alone and menu-driven software package, Bayesian Regression: Nonparametric and Parametric Models, constructed from MATLAB Compiler. Currently, this package gives the user a choice from 83 Bayesian models for data analysis. They include 47 Bayesian nonparametric (BNP) infinite-mixture regression models; 5 BNP infinite-mixture models for density estimation; and 31 normal random effects models (HLMs), including normal linear models. Each of the 78 regression models handles either a continuous, binary, or ordinal dependent variable, and can handle multi-level (grouped) data. All 83 Bayesian models can handle the analysis of weighted observations (e.g., for meta-analysis), and the analysis of left-censored, right-censored, and/or interval-censored data. Each BNP infinite-mixture model has a mixture distribution assigned one of various BNP prior distributions, including priors defined by either the Dirichlet process, Pitman-Yor process (including the normalized stable process), beta (two-parameter) process, normalized inverse-Gaussian process, geometric weights prior, dependent Dirichlet process, or the dependent infinite-probits prior. The software user can mouse-click to select a Bayesian model and perform data analysis via Markov chain Monte Carlo (MCMC) sampling. After the sampling completes, the software automatically opens text output that reports MCMC-based estimates of the model's posterior distribution and model predictive fit to the data. Additional text and/or graphical output can be generated by mouse-clicking other menu options. This includes output of MCMC convergence analyses, and estimates of the model's posterior predictive distribution, for selected functionals and values of covariates. The software is illustrated through the BNP regression analysis of real data.

Modeling groundwater nitrate concentrations in private wells in Iowa

USGS Publications Warehouse

Wheeler, David C.; Nolan, Bernard T.; Flory, Abigail R.; DellaValle, Curt T.; Ward, Mary H.

2015-01-01

Contamination of drinking water by nitrate is a growing problem in many agricultural areas of the country. Ingested nitrate can lead to the endogenous formation of N-nitroso compounds, potent carcinogens. We developed a predictive model for nitrate concentrations in private wells in Iowa. Using 34,084 measurements of nitrate in private wells, we trained and tested random forest models to predict log nitrate levels by systematically assessing the predictive performance of 179 variables in 36 thematic groups (well depth, distance to sinkholes, location, land use, soil characteristics, nitrogen inputs, meteorology, and other factors). The final model contained 66 variables in 17 groups. Some of the most important variables were well depth, slope length within 1 km of the well, year of sample, and distance to nearest animal feeding operation. The correlation between observed and estimated nitrate concentrations was excellent in the training set (r-square = 0.77) and was acceptable in the testing set (r-square = 0.38). The random forest model had substantially better predictive performance than a traditional linear regression model or a regression tree. Our model will be used to investigate the association between nitrate levels in drinking water and cancer risk in the Iowa participants of the Agricultural Health Study cohort.

Modeling groundwater nitrate concentrations in private wells in Iowa.

PubMed

Wheeler, David C; Nolan, Bernard T; Flory, Abigail R; DellaValle, Curt T; Ward, Mary H

2015-12-01

Contamination of drinking water by nitrate is a growing problem in many agricultural areas of the country. Ingested nitrate can lead to the endogenous formation of N-nitroso compounds, potent carcinogens. We developed a predictive model for nitrate concentrations in private wells in Iowa. Using 34,084 measurements of nitrate in private wells, we trained and tested random forest models to predict log nitrate levels by systematically assessing the predictive performance of 179 variables in 36 thematic groups (well depth, distance to sinkholes, location, land use, soil characteristics, nitrogen inputs, meteorology, and other factors). The final model contained 66 variables in 17 groups. Some of the most important variables were well depth, slope length within 1 km of the well, year of sample, and distance to nearest animal feeding operation. The correlation between observed and estimated nitrate concentrations was excellent in the training set (r-square=0.77) and was acceptable in the testing set (r-square=0.38). The random forest model had substantially better predictive performance than a traditional linear regression model or a regression tree. Our model will be used to investigate the association between nitrate levels in drinking water and cancer risk in the Iowa participants of the Agricultural Health Study cohort. Copyright © 2015 Elsevier B.V. All rights reserved.

A regression-based 3-D shoulder rhythm.

PubMed

Xu, Xu; Lin, Jia-hua; McGorry, Raymond W

2014-03-21

In biomechanical modeling of the shoulder, it is important to know the orientation of each bone in the shoulder girdle when estimating the loads on each musculoskeletal element. However, because of the soft tissue overlying the bones, it is difficult to accurately derive the orientation of the clavicle and scapula using surface markers during dynamic movement. The purpose of this study is to develop two regression models which predict the orientation of the clavicle and the scapula. The first regression model uses humerus orientation and individual factors such as age, gender, and anthropometry data as the predictors. The second regression model includes only the humerus orientation as the predictor. Thirty-eight participants performed 118 static postures covering the volume of the right hand reach. The orientation of the thorax, clavicle, scapula and humerus were measured with a motion tracking system. Regression analysis was performed on the Euler angles decomposed from the orientation of each bone from 26 randomly selected participants. The regression models were then validated with the remaining 12 participants. The results indicate that for the first model, the r(2) of the predicted orientation of the clavicle and the scapula ranged between 0.31 and 0.65, and the RMSE obtained from the validation dataset ranged from 6.92° to 10.39°. For the second model, the r(2) ranged between 0.19 and 0.57, and the RMSE obtained from the validation dataset ranged from 6.62° and 11.13°. The derived regression-based shoulder rhythm could be useful in future biomechanical modeling of the shoulder. Copyright © 2014 The Authors. Published by Elsevier Ltd.. All rights reserved.

Characterizing stand-level forest canopy cover and height using Landsat time series, samples of airborne LiDAR, and the Random Forest algorithm

NASA Astrophysics Data System (ADS)

Ahmed, Oumer S.; Franklin, Steven E.; Wulder, Michael A.; White, Joanne C.

2015-03-01

Many forest management activities, including the development of forest inventories, require spatially detailed forest canopy cover and height data. Among the various remote sensing technologies, LiDAR (Light Detection and Ranging) offers the most accurate and consistent means for obtaining reliable canopy structure measurements. A potential solution to reduce the cost of LiDAR data, is to integrate transects (samples) of LiDAR data with frequently acquired and spatially comprehensive optical remotely sensed data. Although multiple regression is commonly used for such modeling, often it does not fully capture the complex relationships between forest structure variables. This study investigates the potential of Random Forest (RF), a machine learning technique, to estimate LiDAR measured canopy structure using a time series of Landsat imagery. The study is implemented over a 2600 ha area of industrially managed coastal temperate forests on Vancouver Island, British Columbia, Canada. We implemented a trajectory-based approach to time series analysis that generates time since disturbance (TSD) and disturbance intensity information for each pixel and we used this information to stratify the forest land base into two strata: mature forests and young forests. Canopy cover and height for three forest classes (i.e. mature, young and mature and young (combined)) were modeled separately using multiple regression and Random Forest (RF) techniques. For all forest classes, the RF models provided improved estimates relative to the multiple regression models. The lowest validation error was obtained for the mature forest strata in a RF model (R2 = 0.88, RMSE = 2.39 m and bias = -0.16 for canopy height; R2 = 0.72, RMSE = 0.068% and bias = -0.0049 for canopy cover). This study demonstrates the value of using disturbance and successional history to inform estimates of canopy structure and obtain improved estimates of forest canopy cover and height using the RF algorithm.

Smooth Scalar-on-Image Regression via Spatial Bayesian Variable Selection

PubMed Central

Goldsmith, Jeff; Huang, Lei; Crainiceanu, Ciprian M.

2013-01-01

We develop scalar-on-image regression models when images are registered multidimensional manifolds. We propose a fast and scalable Bayes inferential procedure to estimate the image coefficient. The central idea is the combination of an Ising prior distribution, which controls a latent binary indicator map, and an intrinsic Gaussian Markov random field, which controls the smoothness of the nonzero coefficients. The model is fit using a single-site Gibbs sampler, which allows fitting within minutes for hundreds of subjects with predictor images containing thousands of locations. The code is simple and is provided in less than one page in the Appendix. We apply this method to a neuroimaging study where cognitive outcomes are regressed on measures of white matter microstructure at every voxel of the corpus callosum for hundreds of subjects. PMID:24729670

Improving satellite-based PM2.5 estimates in China using Gaussian processes modeling in a Bayesian hierarchical setting.

PubMed

Yu, Wenxi; Liu, Yang; Ma, Zongwei; Bi, Jun

2017-08-01

Using satellite-based aerosol optical depth (AOD) measurements and statistical models to estimate ground-level PM 2.5 is a promising way to fill the areas that are not covered by ground PM 2.5 monitors. The statistical models used in previous studies are primarily Linear Mixed Effects (LME) and Geographically Weighted Regression (GWR) models. In this study, we developed a new regression model between PM 2.5 and AOD using Gaussian processes in a Bayesian hierarchical setting. Gaussian processes model the stochastic nature of the spatial random effects, where the mean surface and the covariance function is specified. The spatial stochastic process is incorporated under the Bayesian hierarchical framework to explain the variation of PM 2.5 concentrations together with other factors, such as AOD, spatial and non-spatial random effects. We evaluate the results of our model and compare them with those of other, conventional statistical models (GWR and LME) by within-sample model fitting and out-of-sample validation (cross validation, CV). The results show that our model possesses a CV result (R 2  = 0.81) that reflects higher accuracy than that of GWR and LME (0.74 and 0.48, respectively). Our results indicate that Gaussian process models have the potential to improve the accuracy of satellite-based PM 2.5 estimates.

Landscape-scale consequences of differential tree mortality from catastrophic wind disturbance in the Amazon.

PubMed

Rifai, Sami W; Urquiza Muñoz, José D; Negrón-Juárez, Robinson I; Ramírez Arévalo, Fredy R; Tello-Espinoza, Rodil; Vanderwel, Mark C; Lichstein, Jeremy W; Chambers, Jeffrey Q; Bohlman, Stephanie A

2016-10-01

Wind disturbance can create large forest blowdowns, which greatly reduces live biomass and adds uncertainty to the strength of the Amazon carbon sink. Observational studies from within the central Amazon have quantified blowdown size and estimated total mortality but have not determined which trees are most likely to die from a catastrophic wind disturbance. Also, the impact of spatial dependence upon tree mortality from wind disturbance has seldom been quantified, which is important because wind disturbance often kills clusters of trees due to large treefalls killing surrounding neighbors. We examine (1) the causes of differential mortality between adult trees from a 300-ha blowdown event in the Peruvian region of the northwestern Amazon, (2) how accounting for spatial dependence affects mortality predictions, and (3) how incorporating both differential mortality and spatial dependence affect the landscape level estimation of necromass produced from the blowdown. Standard regression and spatial regression models were used to estimate how stem diameter, wood density, elevation, and a satellite-derived disturbance metric influenced the probability of tree death from the blowdown event. The model parameters regarding tree characteristics, topography, and spatial autocorrelation of the field data were then used to determine the consequences of non-random mortality for landscape production of necromass through a simulation model. Tree mortality was highly non-random within the blowdown, where tree mortality rates were highest for trees that were large, had low wood density, and were located at high elevation. Of the differential mortality models, the non-spatial models overpredicted necromass, whereas the spatial model slightly underpredicted necromass. When parameterized from the same field data, the spatial regression model with differential mortality estimated only 7.5% more dead trees across the entire blowdown than the random mortality model, yet it estimated 51% greater necromass. We suggest that predictions of forest carbon loss from wind disturbance are sensitive to not only the underlying spatial dependence of observations, but also the biological differences between individuals that promote differential levels of mortality. © 2016 by the Ecological Society of America.

Comparison of Random Forest and Parametric Imputation Models for Imputing Missing Data Using MICE: A CALIBER Study

PubMed Central

Shah, Anoop D.; Bartlett, Jonathan W.; Carpenter, James; Nicholas, Owen; Hemingway, Harry

2014-01-01

Multivariate imputation by chained equations (MICE) is commonly used for imputing missing data in epidemiologic research. The “true” imputation model may contain nonlinearities which are not included in default imputation models. Random forest imputation is a machine learning technique which can accommodate nonlinearities and interactions and does not require a particular regression model to be specified. We compared parametric MICE with a random forest-based MICE algorithm in 2 simulation studies. The first study used 1,000 random samples of 2,000 persons drawn from the 10,128 stable angina patients in the CALIBER database (Cardiovascular Disease Research using Linked Bespoke Studies and Electronic Records; 2001–2010) with complete data on all covariates. Variables were artificially made “missing at random,” and the bias and efficiency of parameter estimates obtained using different imputation methods were compared. Both MICE methods produced unbiased estimates of (log) hazard ratios, but random forest was more efficient and produced narrower confidence intervals. The second study used simulated data in which the partially observed variable depended on the fully observed variables in a nonlinear way. Parameter estimates were less biased using random forest MICE, and confidence interval coverage was better. This suggests that random forest imputation may be useful for imputing complex epidemiologic data sets in which some patients have missing data. PMID:24589914

Comparison of random forest and parametric imputation models for imputing missing data using MICE: a CALIBER study.

PubMed

Shah, Anoop D; Bartlett, Jonathan W; Carpenter, James; Nicholas, Owen; Hemingway, Harry

2014-03-15

Multivariate imputation by chained equations (MICE) is commonly used for imputing missing data in epidemiologic research. The "true" imputation model may contain nonlinearities which are not included in default imputation models. Random forest imputation is a machine learning technique which can accommodate nonlinearities and interactions and does not require a particular regression model to be specified. We compared parametric MICE with a random forest-based MICE algorithm in 2 simulation studies. The first study used 1,000 random samples of 2,000 persons drawn from the 10,128 stable angina patients in the CALIBER database (Cardiovascular Disease Research using Linked Bespoke Studies and Electronic Records; 2001-2010) with complete data on all covariates. Variables were artificially made "missing at random," and the bias and efficiency of parameter estimates obtained using different imputation methods were compared. Both MICE methods produced unbiased estimates of (log) hazard ratios, but random forest was more efficient and produced narrower confidence intervals. The second study used simulated data in which the partially observed variable depended on the fully observed variables in a nonlinear way. Parameter estimates were less biased using random forest MICE, and confidence interval coverage was better. This suggests that random forest imputation may be useful for imputing complex epidemiologic data sets in which some patients have missing data.

Influence of therapist competence and quantity of cognitive behavioural therapy on suicidal behaviour and inpatient hospitalisation in a randomised controlled trial in borderline personality disorder: Further analyses of treatment effects in the BOSCOT study

PubMed Central

Norrie, John; Davidson, Kate; Tata, Philip; Gumley, Andrew

2013-01-01

Objectives We investigated the treatment effects reported from a high-quality randomized controlled trial of cognitive behavioural therapy (CBT) for 106 people with borderline personality disorder attending community-based clinics in the UK National Health Service – the BOSCOT trial. Specifically, we examined whether the amount of therapy and therapist competence had an impact on our primary outcome, the number of suicidal acts†, using instrumental variables regression modelling. Design Randomized controlled trial. Participants from across three sites (London, Glasgow, and Ayrshire/Arran) were randomized equally to CBT for personality disorders (CBTpd) plus Treatment as Usual or to Treatment as Usual. Treatment as Usual varied between sites and individuals, but was consistent with routine treatment in the UK National Health Service at the time. CBTpd comprised an average 16 sessions (range 0–35) over 12 months. Method We used instrumental variable regression modelling to estimate the impact of quantity and quality of therapy received (recording activities and behaviours that took place after randomization) on number of suicidal acts and inpatient psychiatric hospitalization. Results A total of 101 participants provided full outcome data at 2 years post randomization. The previously reported intention-to-treat (ITT) results showed on average a reduction of 0.91 (95% confidence interval 0.15–1.67) suicidal acts over 2 years for those randomized to CBT. By incorporating the influence of quantity of therapy and therapist competence, we show that this estimate of the effect of CBTpd could be approximately two to three times greater for those receiving the right amount of therapy from a competent therapist. Conclusions Trials should routinely control for and collect data on both quantity of therapy and therapist competence, which can be used, via instrumental variable regression modelling, to estimate treatment effects for optimal delivery of therapy. Such estimates complement rather than replace the ITT results, which are properly the principal analysis results from such trials. Practitioner points Assessing the impact of the quantity and quality of therapy (competence of therapists) is complex. More competent therapists, trained in CBTpd, may significantly reduce the number of suicidal act in patients with borderline personality disorder. PMID:23420622

Revisiting crash spatial heterogeneity: A Bayesian spatially varying coefficients approach.

PubMed

Xu, Pengpeng; Huang, Helai; Dong, Ni; Wong, S C

2017-01-01

This study was performed to investigate the spatially varying relationships between crash frequency and related risk factors. A Bayesian spatially varying coefficients model was elaborately introduced as a methodological alternative to simultaneously account for the unstructured and spatially structured heterogeneity of the regression coefficients in predicting crash frequencies. The proposed method was appealing in that the parameters were modeled via a conditional autoregressive prior distribution, which involved a single set of random effects and a spatial correlation parameter with extreme values corresponding to pure unstructured or pure spatially correlated random effects. A case study using a three-year crash dataset from the Hillsborough County, Florida, was conducted to illustrate the proposed model. Empirical analysis confirmed the presence of both unstructured and spatially correlated variations in the effects of contributory factors on severe crash occurrences. The findings also suggested that ignoring spatially structured heterogeneity may result in biased parameter estimates and incorrect inferences, while assuming the regression coefficients to be spatially clustered only is probably subject to the issue of over-smoothness. Copyright © 2016 Elsevier Ltd. All rights reserved.

A Comparative Assessment of the Influences of Human Impacts on Soil Cd Concentrations Based on Stepwise Linear Regression, Classification and Regression Tree, and Random Forest Models

PubMed Central

Qiu, Lefeng; Wang, Kai; Long, Wenli; Wang, Ke; Hu, Wei; Amable, Gabriel S.

2016-01-01

Soil cadmium (Cd) contamination has attracted a great deal of attention because of its detrimental effects on animals and humans. This study aimed to develop and compare the performances of stepwise linear regression (SLR), classification and regression tree (CART) and random forest (RF) models in the prediction and mapping of the spatial distribution of soil Cd and to identify likely sources of Cd accumulation in Fuyang County, eastern China. Soil Cd data from 276 topsoil (0–20 cm) samples were collected and randomly divided into calibration (222 samples) and validation datasets (54 samples). Auxiliary data, including detailed land use information, soil organic matter, soil pH, and topographic data, were incorporated into the models to simulate the soil Cd concentrations and further identify the main factors influencing soil Cd variation. The predictive models for soil Cd concentration exhibited acceptable overall accuracies (72.22% for SLR, 70.37% for CART, and 75.93% for RF). The SLR model exhibited the largest predicted deviation, with a mean error (ME) of 0.074 mg/kg, a mean absolute error (MAE) of 0.160 mg/kg, and a root mean squared error (RMSE) of 0.274 mg/kg, and the RF model produced the results closest to the observed values, with an ME of 0.002 mg/kg, an MAE of 0.132 mg/kg, and an RMSE of 0.198 mg/kg. The RF model also exhibited the greatest R2 value (0.772). The CART model predictions closely followed, with ME, MAE, RMSE, and R2 values of 0.013 mg/kg, 0.154 mg/kg, 0.230 mg/kg and 0.644, respectively. The three prediction maps generally exhibited similar and realistic spatial patterns of soil Cd contamination. The heavily Cd-affected areas were primarily located in the alluvial valley plain of the Fuchun River and its tributaries because of the dramatic industrialization and urbanization processes that have occurred there. The most important variable for explaining high levels of soil Cd accumulation was the presence of metal smelting industries. The good performance of the RF model was attributable to its ability to handle the non-linear and hierarchical relationships between soil Cd and environmental variables. These results confirm that the RF approach is promising for the prediction and spatial distribution mapping of soil Cd at the regional scale. PMID:26964095

A Comparative Assessment of the Influences of Human Impacts on Soil Cd Concentrations Based on Stepwise Linear Regression, Classification and Regression Tree, and Random Forest Models.

PubMed

Qiu, Lefeng; Wang, Kai; Long, Wenli; Wang, Ke; Hu, Wei; Amable, Gabriel S

2016-01-01

Soil cadmium (Cd) contamination has attracted a great deal of attention because of its detrimental effects on animals and humans. This study aimed to develop and compare the performances of stepwise linear regression (SLR), classification and regression tree (CART) and random forest (RF) models in the prediction and mapping of the spatial distribution of soil Cd and to identify likely sources of Cd accumulation in Fuyang County, eastern China. Soil Cd data from 276 topsoil (0-20 cm) samples were collected and randomly divided into calibration (222 samples) and validation datasets (54 samples). Auxiliary data, including detailed land use information, soil organic matter, soil pH, and topographic data, were incorporated into the models to simulate the soil Cd concentrations and further identify the main factors influencing soil Cd variation. The predictive models for soil Cd concentration exhibited acceptable overall accuracies (72.22% for SLR, 70.37% for CART, and 75.93% for RF). The SLR model exhibited the largest predicted deviation, with a mean error (ME) of 0.074 mg/kg, a mean absolute error (MAE) of 0.160 mg/kg, and a root mean squared error (RMSE) of 0.274 mg/kg, and the RF model produced the results closest to the observed values, with an ME of 0.002 mg/kg, an MAE of 0.132 mg/kg, and an RMSE of 0.198 mg/kg. The RF model also exhibited the greatest R2 value (0.772). The CART model predictions closely followed, with ME, MAE, RMSE, and R2 values of 0.013 mg/kg, 0.154 mg/kg, 0.230 mg/kg and 0.644, respectively. The three prediction maps generally exhibited similar and realistic spatial patterns of soil Cd contamination. The heavily Cd-affected areas were primarily located in the alluvial valley plain of the Fuchun River and its tributaries because of the dramatic industrialization and urbanization processes that have occurred there. The most important variable for explaining high levels of soil Cd accumulation was the presence of metal smelting industries. The good performance of the RF model was attributable to its ability to handle the non-linear and hierarchical relationships between soil Cd and environmental variables. These results confirm that the RF approach is promising for the prediction and spatial distribution mapping of soil Cd at the regional scale.

Measures of clustering and heterogeneity in multilevel Poisson regression analyses of rates/count data

PubMed Central

Austin, Peter C.; Stryhn, Henrik; Leckie, George; Merlo, Juan

2017-01-01

Multilevel data occur frequently in many research areas like health services research and epidemiology. A suitable way to analyze such data is through the use of multilevel regression models. These models incorporate cluster‐specific random effects that allow one to partition the total variation in the outcome into between‐cluster variation and between‐individual variation. The magnitude of the effect of clustering provides a measure of the general contextual effect. When outcomes are binary or time‐to‐event in nature, the general contextual effect can be quantified by measures of heterogeneity like the median odds ratio or the median hazard ratio, respectively, which can be calculated from a multilevel regression model. Outcomes that are integer counts denoting the number of times that an event occurred are common in epidemiological and medical research. The median (incidence) rate ratio in multilevel Poisson regression for counts that corresponds to the median odds ratio or median hazard ratio for binary or time‐to‐event outcomes respectively is relatively unknown and is rarely used. The median rate ratio is the median relative change in the rate of the occurrence of the event when comparing identical subjects from 2 randomly selected different clusters that are ordered by rate. We also describe how the variance partition coefficient, which denotes the proportion of the variation in the outcome that is attributable to between‐cluster differences, can be computed with count outcomes. We illustrate the application and interpretation of these measures in a case study analyzing the rate of hospital readmission in patients discharged from hospital with a diagnosis of heart failure. PMID:29114926

Bayesian Nonparametric Prediction and Statistical Inference

DTIC Science & Technology

1989-09-07

Kadane, J. (1980), "Bayesian decision theory and the sim- plification of models," in Evaluation of Econometric Models, J. Kmenta and J. Ramsey , eds...the random model and weighted least squares regression," in Evaluation of Econometric Models, ed. by J. Kmenta and J. Ramsey , Academic Press, 197-217...likelihood function. On the other hand, H. Jeffreys’s theory of hypothesis testing covers the most important situations in which the prior is not diffuse. See

Measures of Residual Risk with Connections to Regression, Risk Tracking, Surrogate Models, and Ambiguity

DTIC Science & Technology

2015-04-15

manage , predict, and mitigate the risk in the original variable. Residual risk can be exemplified as a quantification of the improved... the random variable of interest is viewed in concert with a related random vector that helps to manage , predict, and mitigate the risk in the original... manage , predict and mitigate the risk in the original variable. Residual risk can be exemplified as a quantification of the improved situation faced

Fatigue failure of materials under broad band random vibrations

NASA Technical Reports Server (NTRS)

Huang, T. C.; Lanz, R. W.

1971-01-01

The fatigue life of material under multifactor influence of broad band random excitations has been investigated. Parameters which affect the fatigue life are postulated to be peak stress, variance of stress and the natural frequency of the system. Experimental data were processed by the hybrid computer. Based on the experimental results and regression analysis a best predicting model has been found. All values of the experimental fatigue lives are within the 95% confidence intervals of the predicting equation.

Exploring unobserved heterogeneity in bicyclists' red-light running behaviors at different crossing facilities.

PubMed

Guo, Yanyong; Li, Zhibin; Wu, Yao; Xu, Chengcheng

2018-06-01

Bicyclists running the red light at crossing facilities increase the potential of colliding with motor vehicles. Exploring the contributing factors could improve the prediction of running red-light probability and develop countermeasures to reduce such behaviors. However, individuals could have unobserved heterogeneities in running a red light, which make the accurate prediction more challenging. Traditional models assume that factor parameters are fixed and cannot capture the varying impacts on red-light running behaviors. In this study, we employed the full Bayesian random parameters logistic regression approach to account for the unobserved heterogeneous effects. Two types of crossing facilities were considered which were the signalized intersection crosswalks and the road segment crosswalks. Electric and conventional bikes were distinguished in the modeling. Data were collected from 16 crosswalks in urban area of Nanjing, China. Factors such as individual characteristics, road geometric design, environmental features, and traffic variables were examined. Model comparison indicates that the full Bayesian random parameters logistic regression approach is statistically superior to the standard logistic regression model. More red-light runners are predicted at signalized intersection crosswalks than at road segment crosswalks. Factors affecting red-light running behaviors are gender, age, bike type, road width, presence of raised median, separation width, signal type, green ratio, bike and vehicle volume, and average vehicle speed. Factors associated with the unobserved heterogeneity are gender, bike type, signal type, separation width, and bike volume. Copyright © 2018 Elsevier Ltd. All rights reserved.

Novel forecasting approaches using combination of machine learning and statistical models for flood susceptibility mapping.

PubMed

Shafizadeh-Moghadam, Hossein; Valavi, Roozbeh; Shahabi, Himan; Chapi, Kamran; Shirzadi, Ataollah

2018-07-01

In this research, eight individual machine learning and statistical models are implemented and compared, and based on their results, seven ensemble models for flood susceptibility assessment are introduced. The individual models included artificial neural networks, classification and regression trees, flexible discriminant analysis, generalized linear model, generalized additive model, boosted regression trees, multivariate adaptive regression splines, and maximum entropy, and the ensemble models were Ensemble Model committee averaging (EMca), Ensemble Model confidence interval Inferior (EMciInf), Ensemble Model confidence interval Superior (EMciSup), Ensemble Model to estimate the coefficient of variation (EMcv), Ensemble Model to estimate the mean (EMmean), Ensemble Model to estimate the median (EMmedian), and Ensemble Model based on weighted mean (EMwmean). The data set covered 201 flood events in the Haraz watershed (Mazandaran province in Iran) and 10,000 randomly selected non-occurrence points. Among the individual models, the Area Under the Receiver Operating Characteristic (AUROC), which showed the highest value, belonged to boosted regression trees (0.975) and the lowest value was recorded for generalized linear model (0.642). On the other hand, the proposed EMmedian resulted in the highest accuracy (0.976) among all models. In spite of the outstanding performance of some models, nevertheless, variability among the prediction of individual models was considerable. Therefore, to reduce uncertainty, creating more generalizable, more stable, and less sensitive models, ensemble forecasting approaches and in particular the EMmedian is recommended for flood susceptibility assessment. Copyright © 2018 Elsevier Ltd. All rights reserved.

A comparison of different ways of including baseline counts in negative binomial models for data from falls prevention trials.

PubMed

Zheng, Han; Kimber, Alan; Goodwin, Victoria A; Pickering, Ruth M

2018-01-01

A common design for a falls prevention trial is to assess falling at baseline, randomize participants into an intervention or control group, and ask them to record the number of falls they experience during a follow-up period of time. This paper addresses how best to include the baseline count in the analysis of the follow-up count of falls in negative binomial (NB) regression. We examine the performance of various approaches in simulated datasets where both counts are generated from a mixed Poisson distribution with shared random subject effect. Including the baseline count after log-transformation as a regressor in NB regression (NB-logged) or as an offset (NB-offset) resulted in greater power than including the untransformed baseline count (NB-unlogged). Cook and Wei's conditional negative binomial (CNB) model replicates the underlying process generating the data. In our motivating dataset, a statistically significant intervention effect resulted from the NB-logged, NB-offset, and CNB models, but not from NB-unlogged, and large, outlying baseline counts were overly influential in NB-unlogged but not in NB-logged. We conclude that there is little to lose by including the log-transformed baseline count in standard NB regression compared to CNB for moderate to larger sized datasets. © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Predicting Ascospore Release of Monilinia vaccinii-corymbosi of Blueberry with Machine Learning.

PubMed

Harteveld, Dalphy O C; Grant, Michael R; Pscheidt, Jay W; Peever, Tobin L

2017-11-01

Mummy berry, caused by Monilinia vaccinii-corymbosi, causes economic losses of highbush blueberry in the U.S. Pacific Northwest (PNW). Apothecia develop from mummified berries overwintering on soil surfaces and produce ascospores that infect tissue emerging from floral and vegetative buds. Disease control currently relies on fungicides applied on a calendar basis rather than inoculum availability. To establish a prediction model for ascospore release, apothecial development was tracked in three fields, one in western Oregon and two in northwestern Washington in 2015 and 2016. Air and soil temperature, precipitation, soil moisture, leaf wetness, relative humidity and solar radiation were monitored using in-field weather stations and Washington State University's AgWeatherNet stations. Four modeling approaches were compared: logistic regression, multivariate adaptive regression splines, artificial neural networks, and random forest. A supervised learning approach was used to train the models on two data sets: training (70%) and testing (30%). The importance of environmental factors was calculated for each model separately. Soil temperature, soil moisture, and solar radiation were identified as the most important factors influencing ascospore release. Random forest models, with 78% accuracy, showed the best performance compared with the other models. Results of this research helps PNW blueberry growers to optimize fungicide use and reduce production costs.

Modeling and Prediction of Solvent Effect on Human Skin Permeability using Support Vector Regression and Random Forest.

PubMed

Baba, Hiromi; Takahara, Jun-ichi; Yamashita, Fumiyoshi; Hashida, Mitsuru

2015-11-01

The solvent effect on skin permeability is important for assessing the effectiveness and toxicological risk of new dermatological formulations in pharmaceuticals and cosmetics development. The solvent effect occurs by diverse mechanisms, which could be elucidated by efficient and reliable prediction models. However, such prediction models have been hampered by the small variety of permeants and mixture components archived in databases and by low predictive performance. Here, we propose a solution to both problems. We first compiled a novel large database of 412 samples from 261 structurally diverse permeants and 31 solvents reported in the literature. The data were carefully screened to ensure their collection under consistent experimental conditions. To construct a high-performance predictive model, we then applied support vector regression (SVR) and random forest (RF) with greedy stepwise descriptor selection to our database. The models were internally and externally validated. The SVR achieved higher performance statistics than RF. The (externally validated) determination coefficient, root mean square error, and mean absolute error of SVR were 0.899, 0.351, and 0.268, respectively. Moreover, because all descriptors are fully computational, our method can predict as-yet unsynthesized compounds. Our high-performance prediction model offers an attractive alternative to permeability experiments for pharmaceutical and cosmetic candidate screening and optimizing skin-permeable topical formulations.

Teacher Consultation and Coaching within Mental Health Practice: Classroom and Child Effects in Urban Elementary Schools

PubMed Central

Cappella, Elise; Hamre, Bridget K.; Kim, Ha Yeon; Henry, David B.; Frazier, Stacy L.; Atkins, Marc S.; Schoenwald, Sonja K.

2012-01-01

Objective To examine effects of a teacher consultation and coaching program delivered by school and community mental health professionals on change in observed classroom interactions and child functioning across one school year. Method Thirty-six classrooms within five urban elementary schools (87% Latino, 11% Black) were randomly assigned to intervention (training + consultation/coaching) and control (training only) conditions. Classroom and child outcomes (n = 364; 43% girls) were assessed in the fall and spring. Results Random effects regression models showed main effects of intervention on teacher-student relationship closeness, academic self-concept, and peer victimization. Results of multiple regression models showed levels of observed teacher emotional support in the fall moderated intervention impact on emotional support at the end of the school year. Conclusions Results suggest teacher consultation and coaching can be integrated within existing mental health activities in urban schools and impact classroom effectiveness and child adaptation across multiple domains. PMID:22428941

A Global Study of GPP focusing on Light Use Efficiency in a Random Forest Regression Model

NASA Astrophysics Data System (ADS)

Fang, W.; Wei, S.; Yi, C.; Hendrey, G. R.

2016-12-01

Light use efficiency (LUE) is at the core of mechanistic modeling of global gross primary production (GPP). However, most LUE estimates in global models are satellite-based and coarsely measured with emphasis on environmental variables. Others are from eddy covariance towers with much greater spatial and temporal data quality and emphasis on mechanistic processes, but in a limited number of sites. In this paper, we conducted a comprehensive global study of tower-based LUE from 237 FLUXNET towers, and scaled up LUEs from in-situ tower level to global biome level. We integrated key environmental and biological variables into the tower-based LUE estimates, at 0.5o x 0.5o grid-cell resolution, using a random forest regression (RFR) approach. We then developed an RFR-LUE-GPP model using the grid-cell LUE data, and compared it to a tower-LUE-GPP model by the conventional way of treating LUE as a series of biome-specific constants. In order to calibrate the LUE models, we developed a data-driven RFR-GPP model using a random forest regression method. Our results showed that LUE varies largely with latitude. We estimated a global area-weighted average of LUE at 1.21 gC m-2 MJ-1 APAR, which led to an estimated global GPP of 102.9 Gt C /year from 2000 to 2005. The tower-LUE-GPP model tended to overestimate forest GPP in tropical and boreal regions. Large uncertainties exist in GPP estimates over sparsely vegetated areas covered by savannas and woody savannas around the middle to low latitudes (i.g. 20oS to 40oS and 5oN to 15oN) due to lack of available data. Model results were improved by incorporating Köppen climate types to represent climate /meteorological information in machine learning modeling. This shed new light on the recognized issues of climate dependence of spring onset of photosynthesis and the challenges in modeling the biome GPP of evergreen broad leaf forests (EBF) accurately. The divergent responses of GPP to temperature and precipitation at mid-high latitudes and at mid-low latitudes echoed the necessity of modeling GPP separately by latitudes. This work provided a global distribution of LUE estimate, and developed a comprehensive algorithm modeling global terrestrial carbon with high spatial and temporal resolutions.

Using Classification and Regression Trees (CART) and random forests to analyze attrition: Results from two simulations.

PubMed

Hayes, Timothy; Usami, Satoshi; Jacobucci, Ross; McArdle, John J

2015-12-01

In this article, we describe a recent development in the analysis of attrition: using classification and regression trees (CART) and random forest methods to generate inverse sampling weights. These flexible machine learning techniques have the potential to capture complex nonlinear, interactive selection models, yet to our knowledge, their performance in the missing data analysis context has never been evaluated. To assess the potential benefits of these methods, we compare their performance with commonly employed multiple imputation and complete case techniques in 2 simulations. These initial results suggest that weights computed from pruned CART analyses performed well in terms of both bias and efficiency when compared with other methods. We discuss the implications of these findings for applied researchers. (c) 2015 APA, all rights reserved).

Using Classification and Regression Trees (CART) and Random Forests to Analyze Attrition: Results From Two Simulations

PubMed Central

Hayes, Timothy; Usami, Satoshi; Jacobucci, Ross; McArdle, John J.

2016-01-01

In this article, we describe a recent development in the analysis of attrition: using classification and regression trees (CART) and random forest methods to generate inverse sampling weights. These flexible machine learning techniques have the potential to capture complex nonlinear, interactive selection models, yet to our knowledge, their performance in the missing data analysis context has never been evaluated. To assess the potential benefits of these methods, we compare their performance with commonly employed multiple imputation and complete case techniques in 2 simulations. These initial results suggest that weights computed from pruned CART analyses performed well in terms of both bias and efficiency when compared with other methods. We discuss the implications of these findings for applied researchers. PMID:26389526

Association between response rates and survival outcomes in patients with newly diagnosed multiple myeloma. A systematic review and meta-regression analysis.

PubMed

Mainou, Maria; Madenidou, Anastasia-Vasiliki; Liakos, Aris; Paschos, Paschalis; Karagiannis, Thomas; Bekiari, Eleni; Vlachaki, Efthymia; Wang, Zhen; Murad, Mohammad Hassan; Kumar, Shaji; Tsapas, Apostolos

2017-06-01

We performed a systematic review and meta-regression analysis of randomized control trials to investigate the association between response to initial treatment and survival outcomes in patients with newly diagnosed multiple myeloma (MM). Response outcomes included complete response (CR) and the combined outcome of CR or very good partial response (VGPR), while survival outcomes were overall survival (OS) and progression-free survival (PFS). We used random-effect meta-regression models and conducted sensitivity analyses based on definition of CR and study quality. Seventy-two trials were included in the systematic review, 63 of which contributed data in meta-regression analyses. There was no association between OS and CR in patients without autologous stem cell transplant (ASCT) (regression coefficient: .02, 95% confidence interval [CI] -0.06, 0.10), in patients undergoing ASCT (-.11, 95% CI -0.44, 0.22) and in trials comparing ASCT with non-ASCT patients (.04, 95% CI -0.29, 0.38). Similarly, OS did not correlate with the combined metric of CR or VGPR, and no association was evident between response outcomes and PFS. Sensitivity analyses yielded similar results. This meta-regression analysis suggests that there is no association between conventional response outcomes and survival in patients with newly diagnosed MM. © 2017 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.

Security of statistical data bases: invasion of privacy through attribute correlational modeling

DOE Office of Scientific and Technical Information (OSTI.GOV)

Palley, M.A.

This study develops, defines, and applies a statistical technique for the compromise of confidential information in a statistical data base. Attribute Correlational Modeling (ACM) recognizes that the information contained in a statistical data base represents real world statistical phenomena. As such, ACM assumes correlational behavior among the database attributes. ACM proceeds to compromise confidential information through creation of a regression model, where the confidential attribute is treated as the dependent variable. The typical statistical data base may preclude the direct application of regression. In this scenario, the research introduces the notion of a synthetic data base, created through legitimate queriesmore » of the actual data base, and through proportional random variation of responses to these queries. The synthetic data base is constructed to resemble the actual data base as closely as possible in a statistical sense. ACM then applies regression analysis to the synthetic data base, and utilizes the derived model to estimate confidential information in the actual database.« less

A quasi-Monte-Carlo comparison of parametric and semiparametric regression methods for heavy-tailed and non-normal data: an application to healthcare costs.

PubMed

Jones, Andrew M; Lomas, James; Moore, Peter T; Rice, Nigel

2016-10-01

We conduct a quasi-Monte-Carlo comparison of the recent developments in parametric and semiparametric regression methods for healthcare costs, both against each other and against standard practice. The population of English National Health Service hospital in-patient episodes for the financial year 2007-2008 (summed for each patient) is randomly divided into two equally sized subpopulations to form an estimation set and a validation set. Evaluating out-of-sample using the validation set, a conditional density approximation estimator shows considerable promise in forecasting conditional means, performing best for accuracy of forecasting and among the best four for bias and goodness of fit. The best performing model for bias is linear regression with square-root-transformed dependent variables, whereas a generalized linear model with square-root link function and Poisson distribution performs best in terms of goodness of fit. Commonly used models utilizing a log-link are shown to perform badly relative to other models considered in our comparison.

Mixed effect Poisson log-linear models for clinical and epidemiological sleep hypnogram data

PubMed Central

Swihart, Bruce J.; Caffo, Brian S.; Crainiceanu, Ciprian; Punjabi, Naresh M.

2013-01-01

Bayesian Poisson log-linear multilevel models scalable to epidemiological studies are proposed to investigate population variability in sleep state transition rates. Hierarchical random effects are used to account for pairings of subjects and repeated measures within those subjects, as comparing diseased to non-diseased subjects while minimizing bias is of importance. Essentially, non-parametric piecewise constant hazards are estimated and smoothed, allowing for time-varying covariates and segment of the night comparisons. The Bayesian Poisson regression is justified through a re-derivation of a classical algebraic likelihood equivalence of Poisson regression with a log(time) offset and survival regression assuming exponentially distributed survival times. Such re-derivation allows synthesis of two methods currently used to analyze sleep transition phenomena: stratified multi-state proportional hazards models and log-linear models with GEE for transition counts. An example data set from the Sleep Heart Health Study is analyzed. Supplementary material includes the analyzed data set as well as the code for a reproducible analysis. PMID:22241689

A new multiple regression model to identify multi-family houses with a high prevalence of sick building symptoms "SBS", within the healthy sustainable house study in Stockholm (3H).

PubMed

Engvall, Karin; Hult, M; Corner, R; Lampa, E; Norbäck, D; Emenius, G

2010-01-01

The aim was to develop a new model to identify residential buildings with higher frequencies of "SBS" than expected, "risk buildings". In 2005, 481 multi-family buildings with 10,506 dwellings in Stockholm were studied by a new stratified random sampling. A standardised self-administered questionnaire was used to assess "SBS", atopy and personal factors. The response rate was 73%. Statistical analysis was performed by multiple logistic regressions. Dwellers owning their building reported less "SBS" than those renting. There was a strong relationship between socio-economic factors and ownership. The regression model, ended up with high explanatory values for age, gender, atopy and ownership. Applying our model, 9% of all residential buildings in Stockholm were classified as "risk buildings" with the highest proportion in houses built 1961-1975 (26%) and lowest in houses built 1985-1990 (4%). To identify "risk buildings", it is necessary to adjust for ownership and population characteristics.

Predictors of employment status of treated patients with DSM-III-R diagnosis. Can logistic regression model find a solution?

PubMed

Daradkeh, T K; Karim, L

1994-01-01

To investigate the predictors of employment status of patients with DSM-III-R diagnosis, 55 patients were selected by a simple random technique from the main psychiatric clinic in Al Ain, United Arab Emirates. Structured and formal assessments were carried out to extract the potential predictors of outcome of schizophrenia. Logistic regression model revealed that being married, absence of schizoid personality, free or with minimum symptoms of the illness, later age of onset, and higher educational attainment were the most significant predictors of employment outcome. The implications of the results of this study are discussed in the text.

On Models for Binomial Data with Random Numbers of Trials

PubMed Central

Comulada, W. Scott; Weiss, Robert E.

2010-01-01

Summary A binomial outcome is a count s of the number of successes out of the total number of independent trials n = s + f, where f is a count of the failures. The n are random variables not fixed by design in many studies. Joint modeling of (s, f) can provide additional insight into the science and into the probability π of success that cannot be directly incorporated by the logistic regression model. Observations where n = 0 are excluded from the binomial analysis yet may be important to understanding how π is influenced by covariates. Correlation between s and f may exist and be of direct interest. We propose Bayesian multivariate Poisson models for the bivariate response (s, f), correlated through random effects. We extend our models to the analysis of longitudinal and multivariate longitudinal binomial outcomes. Our methodology was motivated by two disparate examples, one from teratology and one from an HIV tertiary intervention study. PMID:17688514

A use of regression analysis in acoustical diagnostics of gear drives

NASA Technical Reports Server (NTRS)

Balitskiy, F. Y.; Genkin, M. D.; Ivanova, M. A.; Kobrinskiy, A. A.; Sokolova, A. G.

1973-01-01

A study is presented of components of the vibration spectrum as the filtered first and second harmonics of the tooth frequency which permits information to be obtained on the physical characteristics of the vibration excitation process, and an approach to be made to comparison of models of the gearing. Regression analysis of two random processes has shown a strong dependence of the second harmonic on the first, and independence of the first from the second. The nature of change in the regression line, with change in loading moment, gives rise to the idea of a variable phase shift between the first and second harmonics.

Application of XGBoost algorithm in hourly PM2.5 concentration prediction

NASA Astrophysics Data System (ADS)

Pan, Bingyue

2018-02-01

In view of prediction techniques of hourly PM2.5 concentration in China, this paper applied the XGBoost(Extreme Gradient Boosting) algorithm to predict hourly PM2.5 concentration. The monitoring data of air quality in Tianjin city was analyzed by using XGBoost algorithm. The prediction performance of the XGBoost method is evaluated by comparing observed and predicted PM2.5 concentration using three measures of forecast accuracy. The XGBoost method is also compared with the random forest algorithm, multiple linear regression, decision tree regression and support vector machines for regression models using computational results. The results demonstrate that the XGBoost algorithm outperforms other data mining methods.

Predicting attention-deficit/hyperactivity disorder severity from psychosocial stress and stress-response genes: a random forest regression approach

PubMed Central

van der Meer, D; Hoekstra, P J; van Donkelaar, M; Bralten, J; Oosterlaan, J; Heslenfeld, D; Faraone, S V; Franke, B; Buitelaar, J K; Hartman, C A

2017-01-01

Identifying genetic variants contributing to attention-deficit/hyperactivity disorder (ADHD) is complicated by the involvement of numerous common genetic variants with small effects, interacting with each other as well as with environmental factors, such as stress exposure. Random forest regression is well suited to explore this complexity, as it allows for the analysis of many predictors simultaneously, taking into account any higher-order interactions among them. Using random forest regression, we predicted ADHD severity, measured by Conners’ Parent Rating Scales, from 686 adolescents and young adults (of which 281 were diagnosed with ADHD). The analysis included 17 374 single-nucleotide polymorphisms (SNPs) across 29 genes previously linked to hypothalamic–pituitary–adrenal (HPA) axis activity, together with information on exposure to 24 individual long-term difficulties or stressful life events. The model explained 12.5% of variance in ADHD severity. The most important SNP, which also showed the strongest interaction with stress exposure, was located in a region regulating the expression of telomerase reverse transcriptase (TERT). Other high-ranking SNPs were found in or near NPSR1, ESR1, GABRA6, PER3, NR3C2 and DRD4. Chronic stressors were more influential than single, severe, life events. Top hits were partly shared with conduct problems. We conclude that random forest regression may be used to investigate how multiple genetic and environmental factors jointly contribute to ADHD. It is able to implicate novel SNPs of interest, interacting with stress exposure, and may explain inconsistent findings in ADHD genetics. This exploratory approach may be best combined with more hypothesis-driven research; top predictors and their interactions with one another should be replicated in independent samples. PMID:28585928

Harmonic regression of Landsat time series for modeling attributes from national forest inventory data

NASA Astrophysics Data System (ADS)

Wilson, Barry T.; Knight, Joseph F.; McRoberts, Ronald E.

2018-03-01

Imagery from the Landsat Program has been used frequently as a source of auxiliary data for modeling land cover, as well as a variety of attributes associated with tree cover. With ready access to all scenes in the archive since 2008 due to the USGS Landsat Data Policy, new approaches to deriving such auxiliary data from dense Landsat time series are required. Several methods have previously been developed for use with finer temporal resolution imagery (e.g. AVHRR and MODIS), including image compositing and harmonic regression using Fourier series. The manuscript presents a study, using Minnesota, USA during the years 2009-2013 as the study area and timeframe. The study examined the relative predictive power of land cover models, in particular those related to tree cover, using predictor variables based solely on composite imagery versus those using estimated harmonic regression coefficients. The study used two common non-parametric modeling approaches (i.e. k-nearest neighbors and random forests) for fitting classification and regression models of multiple attributes measured on USFS Forest Inventory and Analysis plots using all available Landsat imagery for the study area and timeframe. The estimated Fourier coefficients developed by harmonic regression of tasseled cap transformation time series data were shown to be correlated with land cover, including tree cover. Regression models using estimated Fourier coefficients as predictor variables showed a two- to threefold increase in explained variance for a small set of continuous response variables, relative to comparable models using monthly image composites. Similarly, the overall accuracies of classification models using the estimated Fourier coefficients were approximately 10-20 percentage points higher than the models using the image composites, with corresponding individual class accuracies between six and 45 percentage points higher.

Genetic parameters of legendre polynomials for first parity lactation curves.

PubMed

Pool, M H; Janss, L L; Meuwissen, T H

2000-11-01

Variance components of the covariance function coefficients in a random regression test-day model were estimated by Legendre polynomials up to a fifth order for first-parity records of Dutch dairy cows using Gibbs sampling. Two Legendre polynomials of equal order were used to model the random part of the lactation curve, one for the genetic component and one for permanent environment. Test-day records from cows registered between 1990 to 1996 and collected by regular milk recording were available. For the data set, 23,700 complete lactations were selected from 475 herds sired by 262 sires. Because the application of a random regression model is limited by computing capacity, we investigated the minimum order needed to fit the variance structure in the data sufficiently. Predictions of genetic and permanent environmental variance structures were compared with bivariate estimates on 30-d intervals. A third-order or higher polynomial modeled the shape of variance curves over DIM with sufficient accuracy for the genetic and permanent environment part. Also, the genetic correlation structure was fitted with sufficient accuracy by a third-order polynomial, but, for the permanent environmental component, a fourth order was needed. Because equal orders are suggested in the literature, a fourth-order Legendre polynomial is recommended in this study. However, a rank of three for the genetic covariance matrix and of four for permanent environment allows a simpler covariance function with a reduced number of parameters based on the eigenvalues and eigenvectors.

Genetic analysis of groups of mid-infrared predicted fatty acids in milk.

PubMed

Narayana, S G; Schenkel, F S; Fleming, A; Koeck, A; Malchiodi, F; Jamrozik, J; Johnston, J; Sargolzaei, M; Miglior, F

2017-06-01

The objective of this study was to investigate genetic variability of mid-infrared predicted fatty acid groups in Canadian Holstein cattle. Genetic parameters were estimated for 5 groups of fatty acids: short-chain (4 to 10 carbons), medium-chain (11 to 16 carbons), long-chain (17 to 22 carbons), saturated, and unsaturated fatty acids. The data set included 49,127 test-day records from 10,029 first-lactation Holstein cows in 810 herds. The random regression animal test-day model included days in milk, herd-test date, and age-season of calving (polynomial regression) as fixed effects, herd-year of calving, animal additive genetic effect, and permanent environment effects as random polynomial regressions, and random residual effect. Legendre polynomials of the third degree were selected for the fixed regression for age-season of calving effect and Legendre polynomials of the fourth degree were selected for the random regression for animal additive genetic, permanent environment, and herd-year effect. The average daily heritability over the lactation for the medium-chain fatty acid group (0.32) was higher than for the short-chain (0.24) and long-chain (0.23) fatty acid groups. The average daily heritability for the saturated fatty acid group (0.33) was greater than for the unsaturated fatty acid group (0.21). Estimated average daily genetic correlations were positive among all fatty acid groups and ranged from moderate to high (0.63-0.96). The genetic correlations illustrated similarities and differences in their origin and the makeup of the groupings based on chain length and saturation. These results provide evidence for the existence of genetic variation in mid-infrared predicted fatty acid groups, and the possibility of improving milk fatty acid profile through genetic selection in Canadian dairy cattle. Copyright © 2017 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

Testing a single regression coefficient in high dimensional linear models

PubMed Central

Zhong, Ping-Shou; Li, Runze; Wang, Hansheng; Tsai, Chih-Ling

2017-01-01

In linear regression models with high dimensional data, the classical z-test (or t-test) for testing the significance of each single regression coefficient is no longer applicable. This is mainly because the number of covariates exceeds the sample size. In this paper, we propose a simple and novel alternative by introducing the Correlated Predictors Screening (CPS) method to control for predictors that are highly correlated with the target covariate. Accordingly, the classical ordinary least squares approach can be employed to estimate the regression coefficient associated with the target covariate. In addition, we demonstrate that the resulting estimator is consistent and asymptotically normal even if the random errors are heteroscedastic. This enables us to apply the z-test to assess the significance of each covariate. Based on the p-value obtained from testing the significance of each covariate, we further conduct multiple hypothesis testing by controlling the false discovery rate at the nominal level. Then, we show that the multiple hypothesis testing achieves consistent model selection. Simulation studies and empirical examples are presented to illustrate the finite sample performance and the usefulness of the proposed method, respectively. PMID:28663668

Testing a single regression coefficient in high dimensional linear models.

PubMed

Lan, Wei; Zhong, Ping-Shou; Li, Runze; Wang, Hansheng; Tsai, Chih-Ling

2016-11-01

In linear regression models with high dimensional data, the classical z -test (or t -test) for testing the significance of each single regression coefficient is no longer applicable. This is mainly because the number of covariates exceeds the sample size. In this paper, we propose a simple and novel alternative by introducing the Correlated Predictors Screening (CPS) method to control for predictors that are highly correlated with the target covariate. Accordingly, the classical ordinary least squares approach can be employed to estimate the regression coefficient associated with the target covariate. In addition, we demonstrate that the resulting estimator is consistent and asymptotically normal even if the random errors are heteroscedastic. This enables us to apply the z -test to assess the significance of each covariate. Based on the p -value obtained from testing the significance of each covariate, we further conduct multiple hypothesis testing by controlling the false discovery rate at the nominal level. Then, we show that the multiple hypothesis testing achieves consistent model selection. Simulation studies and empirical examples are presented to illustrate the finite sample performance and the usefulness of the proposed method, respectively.

Accelerated Changes in Cortical Thickness Measurements with Age in Military Service Members with Traumatic Brain Injury.

PubMed

Savjani, Ricky R; Taylor, Brian A; Acion, Laura; Wilde, Elisabeth A; Jorge, Ricardo E

2017-11-15

Finding objective and quantifiable imaging markers of mild traumatic brain injury (TBI) has proven challenging, especially in the military population. Changes in cortical thickness after injury have been reported in animals and in humans, but it is unclear how these alterations manifest in the chronic phase, and it is difficult to characterize accurately with imaging. We used cortical thickness measures derived from Advanced Normalization Tools (ANTs) to predict a continuous demographic variable: age. We trained four different regression models (linear regression, support vector regression, Gaussian process regression, and random forests) to predict age from healthy control brains from publicly available datasets (n = 762). We then used these models to predict brain age in military Service Members with TBI (n = 92) and military Service Members without TBI (n = 34). Our results show that all four models overpredicted age in Service Members with TBI, and the predicted age difference was significantly greater compared with military controls. These data extend previous civilian findings and show that cortical thickness measures may reveal an association of accelerated changes over time with military TBI.

Accounting for center in the Early External Cephalic Version trials: an empirical comparison of statistical methods to adjust for center in a multicenter trial with binary outcomes.

PubMed

Reitsma, Angela; Chu, Rong; Thorpe, Julia; McDonald, Sarah; Thabane, Lehana; Hutton, Eileen

2014-09-26

Clustering of outcomes at centers involved in multicenter trials is a type of center effect. The Consolidated Standards of Reporting Trials Statement recommends that multicenter randomized controlled trials (RCTs) should account for center effects in their analysis, however most do not. The Early External Cephalic Version (EECV) trials published in 2003 and 2011 stratified by center at randomization, but did not account for center in the analyses, and due to the nature of the intervention and number of centers, may have been prone to center effects. Using data from the EECV trials, we undertook an empirical study to compare various statistical approaches to account for center effect while estimating the impact of external cephalic version timing (early or delayed) on the outcomes of cesarean section, preterm birth, and non-cephalic presentation at the time of birth. The data from the EECV pilot trial and the EECV2 trial were merged into one dataset. Fisher's exact method was used to test the overall effect of external cephalic version timing unadjusted for center effects. Seven statistical models that accounted for center effects were applied to the data. The models included: i) the Mantel-Haenszel test, ii) logistic regression with fixed center effect and fixed treatment effect, iii) center-size weighted and iv) un-weighted logistic regression with fixed center effect and fixed treatment-by-center interaction, iv) logistic regression with random center effect and fixed treatment effect, v) logistic regression with random center effect and random treatment-by-center interaction, and vi) generalized estimating equations. For each of the three outcomes of interest approaches to account for center effect did not alter the overall findings of the trial. The results were similar for the majority of the methods used to adjust for center, illustrating the robustness of the findings. Despite literature that suggests center effect can change the estimate of effect in multicenter trials, this empirical study does not show a difference in the outcomes of the EECV trials when accounting for center effect. The EECV2 trial was registered on 30 July 30 2005 with Current Controlled Trials: ISRCTN 56498577.

Graphical Models for Quasi-Experimental Designs

ERIC Educational Resources Information Center

Steiner, Peter M.; Kim, Yongnam; Hall, Courtney E.; Su, Dan

2017-01-01

Randomized controlled trials (RCTs) and quasi-experimental designs like regression discontinuity (RD) designs, instrumental variable (IV) designs, and matching and propensity score (PS) designs are frequently used for inferring causal effects. It is well known that the features of these designs facilitate the identification of a causal estimand…

Tehran Air Pollutants Prediction Based on Random Forest Feature Selection Method

NASA Astrophysics Data System (ADS)

Shamsoddini, A.; Aboodi, M. R.; Karami, J.

2017-09-01

Air pollution as one of the most serious forms of environmental pollutions poses huge threat to human life. Air pollution leads to environmental instability, and has harmful and undesirable effects on the environment. Modern prediction methods of the pollutant concentration are able to improve decision making and provide appropriate solutions. This study examines the performance of the Random Forest feature selection in combination with multiple-linear regression and Multilayer Perceptron Artificial Neural Networks methods, in order to achieve an efficient model to estimate carbon monoxide and nitrogen dioxide, sulfur dioxide and PM2.5 contents in the air. The results indicated that Artificial Neural Networks fed by the attributes selected by Random Forest feature selection method performed more accurate than other models for the modeling of all pollutants. The estimation accuracy of sulfur dioxide emissions was lower than the other air contaminants whereas the nitrogen dioxide was predicted more accurate than the other pollutants.

An iteratively reweighted least-squares approach to adaptive robust adjustment of parameters in linear regression models with autoregressive and t-distributed deviations

NASA Astrophysics Data System (ADS)

Kargoll, Boris; Omidalizarandi, Mohammad; Loth, Ina; Paffenholz, Jens-André; Alkhatib, Hamza

2018-03-01

In this paper, we investigate a linear regression time series model of possibly outlier-afflicted observations and autocorrelated random deviations. This colored noise is represented by a covariance-stationary autoregressive (AR) process, in which the independent error components follow a scaled (Student's) t-distribution. This error model allows for the stochastic modeling of multiple outliers and for an adaptive robust maximum likelihood (ML) estimation of the unknown regression and AR coefficients, the scale parameter, and the degree of freedom of the t-distribution. This approach is meant to be an extension of known estimators, which tend to focus only on the regression model, or on the AR error model, or on normally distributed errors. For the purpose of ML estimation, we derive an expectation conditional maximization either algorithm, which leads to an easy-to-implement version of iteratively reweighted least squares. The estimation performance of the algorithm is evaluated via Monte Carlo simulations for a Fourier as well as a spline model in connection with AR colored noise models of different orders and with three different sampling distributions generating the white noise components. We apply the algorithm to a vibration dataset recorded by a high-accuracy, single-axis accelerometer, focusing on the evaluation of the estimated AR colored noise model.

A note on variance estimation in random effects meta-regression.

PubMed

Sidik, Kurex; Jonkman, Jeffrey N

2005-01-01

For random effects meta-regression inference, variance estimation for the parameter estimates is discussed. Because estimated weights are used for meta-regression analysis in practice, the assumed or estimated covariance matrix used in meta-regression is not strictly correct, due to possible errors in estimating the weights. Therefore, this note investigates the use of a robust variance estimation approach for obtaining variances of the parameter estimates in random effects meta-regression inference. This method treats the assumed covariance matrix of the effect measure variables as a working covariance matrix. Using an example of meta-analysis data from clinical trials of a vaccine, the robust variance estimation approach is illustrated in comparison with two other methods of variance estimation. A simulation study is presented, comparing the three methods of variance estimation in terms of bias and coverage probability. We find that, despite the seeming suitability of the robust estimator for random effects meta-regression, the improved variance estimator of Knapp and Hartung (2003) yields the best performance among the three estimators, and thus may provide the best protection against errors in the estimated weights.

Dynamic prediction in functional concurrent regression with an application to child growth.

PubMed

Leroux, Andrew; Xiao, Luo; Crainiceanu, Ciprian; Checkley, William

2018-04-15

In many studies, it is of interest to predict the future trajectory of subjects based on their historical data, referred to as dynamic prediction. Mixed effects models have traditionally been used for dynamic prediction. However, the commonly used random intercept and slope model is often not sufficiently flexible for modeling subject-specific trajectories. In addition, there may be useful exposures/predictors of interest that are measured concurrently with the outcome, complicating dynamic prediction. To address these problems, we propose a dynamic functional concurrent regression model to handle the case where both the functional response and the functional predictors are irregularly measured. Currently, such a model cannot be fit by existing software. We apply the model to dynamically predict children's length conditional on prior length, weight, and baseline covariates. Inference on model parameters and subject-specific trajectories is conducted using the mixed effects representation of the proposed model. An extensive simulation study shows that the dynamic functional regression model provides more accurate estimation and inference than existing methods. Methods are supported by fast, flexible, open source software that uses heavily tested smoothing techniques. © 2017 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

Genetic correlations among body condition score, yield, and fertility in first-parity cows estimated by random regression models.

PubMed

Veerkamp, R F; Koenen, E P; De Jong, G

2001-10-01

Twenty type classifiers scored body condition (BCS) of 91,738 first-parity cows from 601 sires and 5518 maternal grandsires. Fertility data during first lactation were extracted for 177,220 cows, of which 67,278 also had a BCS observation, and first-lactation 305-d milk, fat, and protein yields were added for 180,631 cows. Heritabilities and genetic correlations were estimated using a sire-maternal grandsire model. Heritability of BCS was 0.38. Heritabilities for fertility traits were low (0.01 to 0.07), but genetic standard deviations were substantial, 9 d for days to first service and calving interval, 0.25 for number of services, and 5% for first-service conception. Phenotypic correlations between fertility and yield or BCS were small (-0.15 to 0.20). Genetic correlations between yield and all fertility traits were unfavorable (0.37 to 0.74). Genetic correlations with BCS were between -0.4 and -0.6 for calving interval and days to first service. Random regression analysis (RR) showed that correlations changed with days in milk for BCS. Little agreement was found between variances and correlations from RR, and analysis including a single month (mo 1 to 10) of data for BCS, especially during early and late lactation. However, this was due to excluding data from the conventional analysis, rather than due to the polynomials used. RR and a conventional five-traits model where BCS in mo 1, 4, 7, and 10 was treated as a separate traits (plus yield or fertility) gave similar results. Thus a parsimonious random regression model gave more realistic estimates for the (co)variances than a series of bivariate analysis on subsets of the data for BCS. A higher genetic merit for yield has unfavorable effects on fertility, but the genetic correlation suggests that BCS (at some stages of lactation) might help to alleviate the unfavorable effect of selection for higher yield on fertility.

Practical application of cure mixture model for long-term censored survivor data from a withdrawal clinical trial of patients with major depressive disorder.

PubMed

Arano, Ichiro; Sugimoto, Tomoyuki; Hamasaki, Toshimitsu; Ohno, Yuko

2010-04-23

Survival analysis methods such as the Kaplan-Meier method, log-rank test, and Cox proportional hazards regression (Cox regression) are commonly used to analyze data from randomized withdrawal studies in patients with major depressive disorder. However, unfortunately, such common methods may be inappropriate when a long-term censored relapse-free time appears in data as the methods assume that if complete follow-up were possible for all individuals, each would eventually experience the event of interest. In this paper, to analyse data including such a long-term censored relapse-free time, we discuss a semi-parametric cure regression (Cox cure regression), which combines a logistic formulation for the probability of occurrence of an event with a Cox proportional hazards specification for the time of occurrence of the event. In specifying the treatment's effect on disease-free survival, we consider the fraction of long-term survivors and the risks associated with a relapse of the disease. In addition, we develop a tree-based method for the time to event data to identify groups of patients with differing prognoses (cure survival CART). Although analysis methods typically adapt the log-rank statistic for recursive partitioning procedures, the method applied here used a likelihood ratio (LR) test statistic from a fitting of cure survival regression assuming exponential and Weibull distributions for the latency time of relapse. The method is illustrated using data from a sertraline randomized withdrawal study in patients with major depressive disorder. We concluded that Cox cure regression reveals facts on who may be cured, and how the treatment and other factors effect on the cured incidence and on the relapse time of uncured patients, and that cure survival CART output provides easily understandable and interpretable information, useful both in identifying groups of patients with differing prognoses and in utilizing Cox cure regression models leading to meaningful interpretations.

Assessing variation in life-history tactics within a population using mixture regression models: a practical guide for evolutionary ecologists.

PubMed

Hamel, Sandra; Yoccoz, Nigel G; Gaillard, Jean-Michel

2017-05-01

Mixed models are now well-established methods in ecology and evolution because they allow accounting for and quantifying within- and between-individual variation. However, the required normal distribution of the random effects can often be violated by the presence of clusters among subjects, which leads to multi-modal distributions. In such cases, using what is known as mixture regression models might offer a more appropriate approach. These models are widely used in psychology, sociology, and medicine to describe the diversity of trajectories occurring within a population over time (e.g. psychological development, growth). In ecology and evolution, however, these models are seldom used even though understanding changes in individual trajectories is an active area of research in life-history studies. Our aim is to demonstrate the value of using mixture models to describe variation in individual life-history tactics within a population, and hence to promote the use of these models by ecologists and evolutionary ecologists. We first ran a set of simulations to determine whether and when a mixture model allows teasing apart latent clustering, and to contrast the precision and accuracy of estimates obtained from mixture models versus mixed models under a wide range of ecological contexts. We then used empirical data from long-term studies of large mammals to illustrate the potential of using mixture models for assessing within-population variation in life-history tactics. Mixture models performed well in most cases, except for variables following a Bernoulli distribution and when sample size was small. The four selection criteria we evaluated [Akaike information criterion (AIC), Bayesian information criterion (BIC), and two bootstrap methods] performed similarly well, selecting the right number of clusters in most ecological situations. We then showed that the normality of random effects implicitly assumed by evolutionary ecologists when using mixed models was often violated in life-history data. Mixed models were quite robust to this violation in the sense that fixed effects were unbiased at the population level. However, fixed effects at the cluster level and random effects were better estimated using mixture models. Our empirical analyses demonstrated that using mixture models facilitates the identification of the diversity of growth and reproductive tactics occurring within a population. Therefore, using this modelling framework allows testing for the presence of clusters and, when clusters occur, provides reliable estimates of fixed and random effects for each cluster of the population. In the presence or expectation of clusters, using mixture models offers a suitable extension of mixed models, particularly when evolutionary ecologists aim at identifying how ecological and evolutionary processes change within a population. Mixture regression models therefore provide a valuable addition to the statistical toolbox of evolutionary ecologists. As these models are complex and have their own limitations, we provide recommendations to guide future users. © 2016 Cambridge Philosophical Society.

Modeling Heterogeneity in Relationships between Initial Status and Rates of Change: Treating Latent Variable Regression Coefficients as Random Coefficients in a Three-Level Hierarchical Model

ERIC Educational Resources Information Center

Choi, Kilchan; Seltzer, Michael

2010-01-01

In studies of change in education and numerous other fields, interest often centers on how differences in the status of individuals at the start of a period of substantive interest relate to differences in subsequent change. In this article, the authors present a fully Bayesian approach to estimating three-level Hierarchical Models in which latent…

The Development of a Stochastic Model of the Atmosphere Between 30 and 90 Km to Be Used in Determining the Effect of Atmospheric Variability on Space Shuttle Entry Parameters. Ph.D. Thesis - Virginia Polytechnic Inst. and State Univ.

NASA Technical Reports Server (NTRS)

Campbell, J. W.

1973-01-01

A stochasitc model of the atmosphere between 30 and 90 km was developed for use in Monte Carlo space shuttle entry studies. The model is actually a family of models, one for each latitude-season category as defined in the 1966 U.S. Standard Atmosphere Supplements. Each latitude-season model generates a pseudo-random temperature profile whose mean is the appropriate temperature profile from the Standard Atmosphere Supplements. The standard deviation of temperature at each altitude for a given latitude-season model was estimated from sounding-rocket data. Departures from the mean temperature at each altitude were produced by assuming a linear regression of temperature on the solar heating rate of ozone. A profile of random ozone concentrations was first generated using an auxiliary stochastic ozone model, also developed as part of this study, and then solar heating rates were computed for the random ozone concentrations.

Household water treatment in developing countries: comparing different intervention types using meta-regression.

PubMed

Hunter, Paul R

2009-12-01

Household water treatment (HWT) is being widely promoted as an appropriate intervention for reducing the burden of waterborne disease in poor communities in developing countries. A recent study has raised concerns about the effectiveness of HWT, in part because of concerns over the lack of blinding and in part because of considerable heterogeneity in the reported effectiveness of randomized controlled trials. This study set out to attempt to investigate the causes of this heterogeneity and so identify factors associated with good health gains. Studies identified in an earlier systematic review and meta-analysis were supplemented with more recently published randomized controlled trials. A total of 28 separate studies of randomized controlled trials of HWT with 39 intervention arms were included in the analysis. Heterogeneity was studied using the "metareg" command in Stata. Initial analyses with single candidate predictors were undertaken and all variables significant at the P < 0.2 level were included in a final regression model. Further analyses were done to estimate the effect of the interventions over time by MonteCarlo modeling using @Risk and the parameter estimates from the final regression model. The overall effect size of all unblinded studies was relative risk = 0.56 (95% confidence intervals 0.51-0.63), but after adjusting for bias due to lack of blinding the effect size was much lower (RR = 0.85, 95% CI = 0.76-0.97). Four main variables were significant predictors of effectiveness of intervention in a multipredictor meta regression model: Log duration of study follow-up (regression coefficient of log effect size = 0.186, standard error (SE) = 0.072), whether or not the study was blinded (coefficient 0.251, SE 0.066) and being conducted in an emergency setting (coefficient -0.351, SE 0.076) were all significant predictors of effect size in the final model. Compared to the ceramic filter all other interventions were much less effective (Biosand 0.247, 0.073; chlorine and safe waste storage 0.295, 0.061; combined coagulant-chlorine 0.2349, 0.067; SODIS 0.302, 0.068). A Monte Carlo model predicted that over 12 months ceramic filters were likely to be still effective at reducing disease, whereas SODIS, chlorination, and coagulation-chlorination had little if any benefit. Indeed these three interventions are predicted to have the same or less effect than what may be expected due purely to reporting bias in unblinded studies With the currently available evidence ceramic filters are the most effective form of HWT in the longterm, disinfection-only interventions including SODIS appear to have poor if any longterm public health benefit.

Modeling contemporary climate profiles of whitebark pine (Pinus albicaulis) and predicting responses to global warming

Treesearch

Marcus V. Warwell; Gerald E. Rehfeldt; Nicholas L. Crookston

2006-01-01

The Random Forests multiple regression tree was used to develop an empirically-based bioclimate model for the distribution of Pinus albicaulis (whitebark pine) in western North America, latitudes 31° to 51° N and longitudes 102° to 125° W. Independent variables included 35 simple expressions of temperature and precipitation and their interactions....

Genetic structured antedependence and random regression models applied to the longitudinal feed conversion ratio in growing Large White pigs.

PubMed

Huynh-Tran, V H; Gilbert, H; David, I

2017-11-01

The objective of the present study was to compare a random regression model, usually used in genetic analyses of longitudinal data, with the structured antedependence (SAD) model to study the longitudinal feed conversion ratio (FCR) in growing Large White pigs and to propose criteria for animal selection when used for genetic evaluation. The study was based on data from 11,790 weekly FCR measures collected on 1,186 Large White male growing pigs. Random regression (RR) using orthogonal polynomial Legendre and SAD models was used to estimate genetic parameters and predict FCR-based EBV for each of the 10 wk of the test. The results demonstrated that the best SAD model (1 order of antedependence of degree 2 and a polynomial of degree 2 for the innovation variance for the genetic and permanent environmental effects, i.e., 12 parameters) provided a better fit for the data than RR with a quadratic function for the genetic and permanent environmental effects (13 parameters), with Bayesian information criteria values of -10,060 and -9,838, respectively. Heritabilities with the SAD model were higher than those of RR over the first 7 wk of the test. Genetic correlations between weeks were higher than 0.68 for short intervals between weeks and decreased to 0.08 for the SAD model and -0.39 for RR for the longest intervals. These differences in genetic parameters showed that, contrary to the RR approach, the SAD model does not suffer from border effect problems and can handle genetic correlations that tend to 0. Summarized breeding values were proposed for each approach as linear combinations of the individual weekly EBV weighted by the coefficients of the first or second eigenvector computed from the genetic covariance matrix of the additive genetic effects. These summarized breeding values isolated EBV trajectories over time, capturing either the average general value or the slope of the trajectory. Finally, applying the SAD model over a reduced period of time suggested that similar selection choices would result from the use of the records from the first 8 wk of the test. To conclude, the SAD model performed well for the genetic evaluation of longitudinal phenotypes.

Exact and Approximate Statistical Inference for Nonlinear Regression and the Estimating Equation Approach.

PubMed

Demidenko, Eugene

2017-09-01

The exact density distribution of the nonlinear least squares estimator in the one-parameter regression model is derived in closed form and expressed through the cumulative distribution function of the standard normal variable. Several proposals to generalize this result are discussed. The exact density is extended to the estimating equation (EE) approach and the nonlinear regression with an arbitrary number of linear parameters and one intrinsically nonlinear parameter. For a very special nonlinear regression model, the derived density coincides with the distribution of the ratio of two normally distributed random variables previously obtained by Fieller (1932), unlike other approximations previously suggested by other authors. Approximations to the density of the EE estimators are discussed in the multivariate case. Numerical complications associated with the nonlinear least squares are illustrated, such as nonexistence and/or multiple solutions, as major factors contributing to poor density approximation. The nonlinear Markov-Gauss theorem is formulated based on the near exact EE density approximation.

Exposure assessment models for elemental components of particulate matter in an urban environment: A comparison of regression and random forest approaches

NASA Astrophysics Data System (ADS)

Brokamp, Cole; Jandarov, Roman; Rao, M. B.; LeMasters, Grace; Ryan, Patrick

2017-02-01

Exposure assessment for elemental components of particulate matter (PM) using land use modeling is a complex problem due to the high spatial and temporal variations in pollutant concentrations at the local scale. Land use regression (LUR) models may fail to capture complex interactions and non-linear relationships between pollutant concentrations and land use variables. The increasing availability of big spatial data and machine learning methods present an opportunity for improvement in PM exposure assessment models. In this manuscript, our objective was to develop a novel land use random forest (LURF) model and compare its accuracy and precision to a LUR model for elemental components of PM in the urban city of Cincinnati, Ohio. PM smaller than 2.5 μm (PM2.5) and eleven elemental components were measured at 24 sampling stations from the Cincinnati Childhood Allergy and Air Pollution Study (CCAAPS). Over 50 different predictors associated with transportation, physical features, community socioeconomic characteristics, greenspace, land cover, and emission point sources were used to construct LUR and LURF models. Cross validation was used to quantify and compare model performance. LURF and LUR models were created for aluminum (Al), copper (Cu), iron (Fe), potassium (K), manganese (Mn), nickel (Ni), lead (Pb), sulfur (S), silicon (Si), vanadium (V), zinc (Zn), and total PM2.5 in the CCAAPS study area. LURF utilized a more diverse and greater number of predictors than LUR and LURF models for Al, K, Mn, Pb, Si, Zn, TRAP, and PM2.5 all showed a decrease in fractional predictive error of at least 5% compared to their LUR models. LURF models for Al, Cu, Fe, K, Mn, Pb, Si, Zn, TRAP, and PM2.5 all had a cross validated fractional predictive error less than 30%. Furthermore, LUR models showed a differential exposure assessment bias and had a higher prediction error variance. Random forest and other machine learning methods may provide more accurate exposure assessment.

Exposure assessment models for elemental components of particulate matter in an urban environment: A comparison of regression and random forest approaches.

PubMed

Brokamp, Cole; Jandarov, Roman; Rao, M B; LeMasters, Grace; Ryan, Patrick

2017-02-01

Exposure assessment for elemental components of particulate matter (PM) using land use modeling is a complex problem due to the high spatial and temporal variations in pollutant concentrations at the local scale. Land use regression (LUR) models may fail to capture complex interactions and non-linear relationships between pollutant concentrations and land use variables. The increasing availability of big spatial data and machine learning methods present an opportunity for improvement in PM exposure assessment models. In this manuscript, our objective was to develop a novel land use random forest (LURF) model and compare its accuracy and precision to a LUR model for elemental components of PM in the urban city of Cincinnati, Ohio. PM smaller than 2.5 μm (PM2.5) and eleven elemental components were measured at 24 sampling stations from the Cincinnati Childhood Allergy and Air Pollution Study (CCAAPS). Over 50 different predictors associated with transportation, physical features, community socioeconomic characteristics, greenspace, land cover, and emission point sources were used to construct LUR and LURF models. Cross validation was used to quantify and compare model performance. LURF and LUR models were created for aluminum (Al), copper (Cu), iron (Fe), potassium (K), manganese (Mn), nickel (Ni), lead (Pb), sulfur (S), silicon (Si), vanadium (V), zinc (Zn), and total PM2.5 in the CCAAPS study area. LURF utilized a more diverse and greater number of predictors than LUR and LURF models for Al, K, Mn, Pb, Si, Zn, TRAP, and PM2.5 all showed a decrease in fractional predictive error of at least 5% compared to their LUR models. LURF models for Al, Cu, Fe, K, Mn, Pb, Si, Zn, TRAP, and PM2.5 all had a cross validated fractional predictive error less than 30%. Furthermore, LUR models showed a differential exposure assessment bias and had a higher prediction error variance. Random forest and other machine learning methods may provide more accurate exposure assessment.

Exposure assessment models for elemental components of particulate matter in an urban environment: A comparison of regression and random forest approaches

PubMed Central

Brokamp, Cole; Jandarov, Roman; Rao, M.B.; LeMasters, Grace; Ryan, Patrick

2017-01-01

Exposure assessment for elemental components of particulate matter (PM) using land use modeling is a complex problem due to the high spatial and temporal variations in pollutant concentrations at the local scale. Land use regression (LUR) models may fail to capture complex interactions and non-linear relationships between pollutant concentrations and land use variables. The increasing availability of big spatial data and machine learning methods present an opportunity for improvement in PM exposure assessment models. In this manuscript, our objective was to develop a novel land use random forest (LURF) model and compare its accuracy and precision to a LUR model for elemental components of PM in the urban city of Cincinnati, Ohio. PM smaller than 2.5 μm (PM2.5) and eleven elemental components were measured at 24 sampling stations from the Cincinnati Childhood Allergy and Air Pollution Study (CCAAPS). Over 50 different predictors associated with transportation, physical features, community socioeconomic characteristics, greenspace, land cover, and emission point sources were used to construct LUR and LURF models. Cross validation was used to quantify and compare model performance. LURF and LUR models were created for aluminum (Al), copper (Cu), iron (Fe), potassium (K), manganese (Mn), nickel (Ni), lead (Pb), sulfur (S), silicon (Si), vanadium (V), zinc (Zn), and total PM2.5 in the CCAAPS study area. LURF utilized a more diverse and greater number of predictors than LUR and LURF models for Al, K, Mn, Pb, Si, Zn, TRAP, and PM2.5 all showed a decrease in fractional predictive error of at least 5% compared to their LUR models. LURF models for Al, Cu, Fe, K, Mn, Pb, Si, Zn, TRAP, and PM2.5 all had a cross validated fractional predictive error less than 30%. Furthermore, LUR models showed a differential exposure assessment bias and had a higher prediction error variance. Random forest and other machine learning methods may provide more accurate exposure assessment. PMID:28959135

Developing Data-driven models for quantifying Cochlodinium polykrikoides in Coastal Waters

NASA Astrophysics Data System (ADS)

Kwon, Yongsung; Jang, Eunna; Im, Jungho; Baek, Seungho; Park, Yongeun; Cho, Kyunghwa

2017-04-01

Harmful algal blooms have been worldwide problems because it leads to serious dangers to human health and aquatic ecosystems. Especially, fish killing red tide blooms by one of dinoflagellate, Cochlodinium polykrikoides (C. polykrikoides), have caused critical damage to mariculture in the Korean coastal waters. In this work, multiple linear regression (MLR), regression tree (RT), and random forest (RF) models were constructed and applied to estimate C. polykrikoides blooms in coastal waters. Five different types of input dataset were carried out to test the performance of three models. To train and validate the three models, observed number of C. polykrikoides cells from National institute of fisheries science (NIFS) and remote sensing reflectance data from Geostationary Ocean Color Imager (GOCI) images for 3 years from 2013 to 2015 were used. The RT model showed the best prediction performance when using 4 bands and 3 band ratios data were used as input data simultaneously. Results obtained from iterative model development with randomly chosen input data indicated that the recognition of patterns in training data caused a variation in prediction performance. This work provided useful tools for reliably estimate the number of C. polykrikoides cells by using reasonable input reflectance dataset in coastal waters. It is expected that the RT model is easily accessed and manipulated by administrators and decision-makers working with coastal waters.

Genetic interactions for heat stress and production level: predicting foreign from domestic data

USDA-ARS?s Scientific Manuscript database

Genetic by environmental interactions were estimated from U.S. national data by separately adding random regressions for heat stress (HS) and herd production level (HL) to the all-breed animal model to improve predictions of future records and rankings in other climate and production situations. Yie...

A computational approach to compare regression modelling strategies in prediction research.

PubMed

Pajouheshnia, Romin; Pestman, Wiebe R; Teerenstra, Steven; Groenwold, Rolf H H

2016-08-25

It is often unclear which approach to fit, assess and adjust a model will yield the most accurate prediction model. We present an extension of an approach for comparing modelling strategies in linear regression to the setting of logistic regression and demonstrate its application in clinical prediction research. A framework for comparing logistic regression modelling strategies by their likelihoods was formulated using a wrapper approach. Five different strategies for modelling, including simple shrinkage methods, were compared in four empirical data sets to illustrate the concept of a priori strategy comparison. Simulations were performed in both randomly generated data and empirical data to investigate the influence of data characteristics on strategy performance. We applied the comparison framework in a case study setting. Optimal strategies were selected based on the results of a priori comparisons in a clinical data set and the performance of models built according to each strategy was assessed using the Brier score and calibration plots. The performance of modelling strategies was highly dependent on the characteristics of the development data in both linear and logistic regression settings. A priori comparisons in four empirical data sets found that no strategy consistently outperformed the others. The percentage of times that a model adjustment strategy outperformed a logistic model ranged from 3.9 to 94.9 %, depending on the strategy and data set. However, in our case study setting the a priori selection of optimal methods did not result in detectable improvement in model performance when assessed in an external data set. The performance of prediction modelling strategies is a data-dependent process and can be highly variable between data sets within the same clinical domain. A priori strategy comparison can be used to determine an optimal logistic regression modelling strategy for a given data set before selecting a final modelling approach.

Breastfeeding and intelligence: a systematic review and meta-analysis.

PubMed

Horta, Bernardo L; Loret de Mola, Christian; Victora, Cesar G

2015-12-01

This study was aimed at systematically reviewing evidence of the association between breastfeeding and performance in intelligence tests. Two independent searches were carried out using Medline, LILACS, SCIELO and Web of Science. Studies restricted to infants and those where estimates were not adjusted for stimulation or interaction at home were excluded. Fixed- and random-effects models were used to pool the effect estimates, and a random-effects regression was used to assess potential sources of heterogeneity. We included 17 studies with 18 estimates of the relationship between breastfeeding and performance in intelligence tests. In a random-effects model, breastfed subjects achieved a higher IQ [mean difference: 3.44 points (95% confidence interval: 2.30; 4.58)]. We found no evidence of publication bias. Studies that controlled for maternal IQ showed a smaller benefit from breastfeeding [mean difference 2.62 points (95% confidence interval: 1.25; 3.98)]. In the meta-regression, none of the study characteristics explained the heterogeneity among the studies. Breastfeeding is related to improved performance in intelligence tests. A positive effect of breastfeeding on cognition was also observed in a randomised trial. This suggests that the association is causal. ©2015 The Authors. Acta Paediatrica published by John Wiley & Sons Ltd on behalf of Foundation Acta Paediatrica.

Hyper-Spectral Image Analysis With Partially Latent Regression and Spatial Markov Dependencies

NASA Astrophysics Data System (ADS)

Deleforge, Antoine; Forbes, Florence; Ba, Sileye; Horaud, Radu

2015-09-01

Hyper-spectral data can be analyzed to recover physical properties at large planetary scales. This involves resolving inverse problems which can be addressed within machine learning, with the advantage that, once a relationship between physical parameters and spectra has been established in a data-driven fashion, the learned relationship can be used to estimate physical parameters for new hyper-spectral observations. Within this framework, we propose a spatially-constrained and partially-latent regression method which maps high-dimensional inputs (hyper-spectral images) onto low-dimensional responses (physical parameters such as the local chemical composition of the soil). The proposed regression model comprises two key features. Firstly, it combines a Gaussian mixture of locally-linear mappings (GLLiM) with a partially-latent response model. While the former makes high-dimensional regression tractable, the latter enables to deal with physical parameters that cannot be observed or, more generally, with data contaminated by experimental artifacts that cannot be explained with noise models. Secondly, spatial constraints are introduced in the model through a Markov random field (MRF) prior which provides a spatial structure to the Gaussian-mixture hidden variables. Experiments conducted on a database composed of remotely sensed observations collected from the Mars planet by the Mars Express orbiter demonstrate the effectiveness of the proposed model.

Chain pooling to minimize prediction error in subset regression. [Monte Carlo studies using population models

NASA Technical Reports Server (NTRS)

Holms, A. G.

1974-01-01

Monte Carlo studies using population models intended to represent response surface applications are reported. Simulated experiments were generated by adding pseudo random normally distributed errors to population values to generate observations. Model equations were fitted to the observations and the decision procedure was used to delete terms. Comparison of values predicted by the reduced models with the true population values enabled the identification of deletion strategies that are approximately optimal for minimizing prediction errors.

Improvement of Storm Forecasts Using Gridded Bayesian Linear Regression for Northeast United States

NASA Astrophysics Data System (ADS)

Yang, J.; Astitha, M.; Schwartz, C. S.

2017-12-01

Bayesian linear regression (BLR) is a post-processing technique in which regression coefficients are derived and used to correct raw forecasts based on pairs of observation-model values. This study presents the development and application of a gridded Bayesian linear regression (GBLR) as a new post-processing technique to improve numerical weather prediction (NWP) of rain and wind storm forecasts over northeast United States. Ten controlled variables produced from ten ensemble members of the National Center for Atmospheric Research (NCAR) real-time prediction system are used for a GBLR model. In the GBLR framework, leave-one-storm-out cross-validation is utilized to study the performances of the post-processing technique in a database composed of 92 storms. To estimate the regression coefficients of the GBLR, optimization procedures that minimize the systematic and random error of predicted atmospheric variables (wind speed, precipitation, etc.) are implemented for the modeled-observed pairs of training storms. The regression coefficients calculated for meteorological stations of the National Weather Service are interpolated back to the model domain. An analysis of forecast improvements based on error reductions during the storms will demonstrate the value of GBLR approach. This presentation will also illustrate how the variances are optimized for the training partition in GBLR and discuss the verification strategy for grid points where no observations are available. The new post-processing technique is successful in improving wind speed and precipitation storm forecasts using past event-based data and has the potential to be implemented in real-time.

Evaluation of the Carrying Capacity of Rectangular Steel-Concrete Columns

NASA Astrophysics Data System (ADS)

Vatulia, Glib; Rezunenko, Maryna; Petrenko, Dmytro; Rezunenko, Sergii

2018-06-01

Experimental studies of rectangular steel-concrete columns under centric compression with random eccentricity were conducted. The stress-strain state and the carrying capacity exhaustion have been assessed. The regression dependence is proposed to determine the maximum carrying capacity of such columns. The mathematical model takes into account the combined influence of the physical and geometric characteristics of the columns, such as their length, crosssectional area, casing thickness, prism strength of concrete, yield strength of steel, modulus of elasticity of both steel and concrete. The correspondence of the obtained model to the experimental data, as well as the significance of the regression parameters are confirmed by the Fisher and Student criteria.

The role of gender in a smoking cessation intervention: a cluster randomized clinical trial.

PubMed

Puente, Diana; Cabezas, Carmen; Rodriguez-Blanco, Teresa; Fernández-Alonso, Carmen; Cebrian, Tránsito; Torrecilla, Miguel; Clemente, Lourdes; Martín, Carlos

2011-05-23

The prevalence of smoking in Spain is high in both men and women. The aim of our study was to evaluate the role of gender in the effectiveness of a specific smoking cessation intervention conducted in Spain. This study was a secondary analysis of a cluster randomized clinical trial in which the randomization unit was the Basic Care Unit (family physician and nurse who care for the same group of patients). The intervention consisted of a six-month period of implementing the recommendations of a Clinical Practice Guideline. A total of 2,937 current smokers at 82 Primary Care Centers in 13 different regions of Spain were included (2003-2005). The success rate was measured by a six-month continued abstinence rate at the one-year follow-up. A logistic mixed-effects regression model, taking Basic Care Units as random-effect parameter, was performed in order to analyze gender as a predictor of smoking cessation. At the one-year follow-up, the six-month continuous abstinence quit rate was 9.4% in men and 8.5% in women (p = 0.400). The logistic mixed-effects regression model showed that women did not have a higher odds of being an ex-smoker than men after the analysis was adjusted for confounders (OR adjusted = 0.9, 95% CI = 0.7-1.2). Gender does not appear to be a predictor of smoking cessation at the one-year follow-up in individuals presenting at Primary Care Centers. CLINICALTRIALS.GOV IDENTIFIER: NCT00125905.

Zinc supplementation for the prevention of acute lower respiratory infection in children in developing countries: meta-analysis and meta-regression of randomized trials.

PubMed

Roth, Daniel E; Richard, Stephanie A; Black, Robert E

2010-06-01

Routine zinc supplementation is a potential intervention for the prevention of acute lower respiratory infection (ALRI) in developing countries. However, discrepant findings from recent randomized trials remain unexplained. Randomized trials of zinc supplementation in young children in developing countries were identified by a systematic literature review. Trials included in the meta-analysis met specific criteria, including participants <5 years of age, daily/weekly zinc and control supplementation for greater than 3 months, active household surveillance for respiratory morbidity and use of a case definition that included at least one sign of lower respiratory tract illness. ALRI case definitions were classified on the basis of specificity/severity. Incidence rate ratios (IRRs) were pooled by random-effects models. Meta-regression and sub-group analysis were performed to assess potential sources of between-study heterogeneity. Ten trials were eligible for inclusion (n = 49 450 children randomized). Zinc reduced the incidence of ALRI defined by specific clinical criteria [IRR 0.65, 95% confidence interval (CI) 0.52-0.82], but had no effect on lower-specificity ALRI case definitions based on caregiver report (IRR 1.01, 95% CI 0.91-1.12) or World Health Organization 'non-severe pneumonia' (0.96, 95% CI 0.86-1.08). By meta-regression, the effect of zinc was associated with ALRI case definition, but not with mean baseline age, geographic location, nutritional status or zinc dose. Routine zinc supplementation reduced the incidence of childhood ALRI defined by relatively specific clinical criteria, but the effect was null if lower specificity case definitions were applied. The choice of ALRI case definition may substantially influence inferences from community trials regarding the efficacy of preventive interventions.

Prediction of Short-Distance Aerial Movement of Phakopsora pachyrhizi Urediniospores Using Machine Learning.

PubMed

Wen, L; Bowen, C R; Hartman, G L

2017-10-01

Dispersal of urediniospores by wind is the primary means of spread for Phakopsora pachyrhizi, the cause of soybean rust. Our research focused on the short-distance movement of urediniospores from within the soybean canopy and up to 61 m from field-grown rust-infected soybean plants. Environmental variables were used to develop and compare models including the least absolute shrinkage and selection operator regression, zero-inflated Poisson/regular Poisson regression, random forest, and neural network to describe deposition of urediniospores collected in passive and active traps. All four models identified distance of trap from source, humidity, temperature, wind direction, and wind speed as the five most important variables influencing short-distance movement of urediniospores. The random forest model provided the best predictions, explaining 76.1 and 86.8% of the total variation in the passive- and active-trap datasets, respectively. The prediction accuracy based on the correlation coefficient (r) between predicted values and the true values were 0.83 (P < 0.0001) and 0.94 (P < 0.0001) for the passive and active trap datasets, respectively. Overall, multiple machine learning techniques identified the most important variables to make the most accurate predictions of movement of P. pachyrhizi urediniospores short-distance.

[Using fractional polynomials to estimate the safety threshold of fluoride in drinking water].

PubMed

Pan, Shenling; An, Wei; Li, Hongyan; Yang, Min

2014-01-01

To study the dose-response relationship between fluoride content in drinking water and prevalence of dental fluorosis on the national scale, then to determine the safety threshold of fluoride in drinking water. Meta-regression analysis was applied to the 2001-2002 national endemic fluorosis survey data of key wards. First, fractional polynomial (FP) was adopted to establish fixed effect model, determining the best FP structure, after that restricted maximum likelihood (REML) was adopted to estimate between-study variance, then the best random effect model was established. The best FP structure was first-order logarithmic transformation. Based on the best random effect model, the benchmark dose (BMD) of fluoride in drinking water and its lower limit (BMDL) was calculated as 0.98 mg/L and 0.78 mg/L. Fluoride in drinking water can only explain 35.8% of the variability of the prevalence, among other influencing factors, ward type was a significant factor, while temperature condition and altitude were not. Fractional polynomial-based meta-regression method is simple, practical and can provide good fitting effect, based on it, the safety threshold of fluoride in drinking water of our country is determined as 0.8 mg/L.

Evaluation of genotype x environment interactions in cotton using the method proposed by Eberhart and Russell and reaction norm models.

PubMed

Alves, R S; Teodoro, P E; Farias, F C; Farias, F J C; Carvalho, L P; Rodrigues, J I S; Bhering, L L; Resende, M D V

2017-08-17

Cotton produces one of the most important textile fibers of the world and has great relevance in the world economy. It is an economically important crop in Brazil, which is the world's fifth largest producer. However, studies evaluating the genotype x environment (G x E) interactions in cotton are scarce in this country. Therefore, the goal of this study was to evaluate the G x E interactions in two important traits in cotton (fiber yield and fiber length) using the method proposed by Eberhart and Russell (simple linear regression) and reaction norm models (random regression). Eight trials with sixteen upland cotton genotypes, conducted in a randomized block design, were used. It was possible to identify a genotype with wide adaptability and stability for both traits. Reaction norm models have excellent theoretical and practical properties and led to more informative and accurate results than the method proposed by Eberhart and Russell and should, therefore, be preferred. Curves of genotypic values as a function of the environmental gradient, which predict the behavior of the genotypes along the environmental gradient, were generated. These curves make possible the recommendation to untested environmental levels.

Genetic evaluation of weekly body weight in Japanese quail using random regression models.

PubMed

Karami, K; Zerehdaran, S; Tahmoorespur, M; Barzanooni, B; Lotfi, E

2017-02-01

1. A total of 11 826 records from 2489 quails, hatched between 2012 and 2013, were used to estimate genetic parameters for BW (body weight) of Japanese quail using random regression models. Weekly BW was measured from hatch until 49 d of age. WOMBAT software (University of New England, Australia) was used for estimating genetic and phenotypic parameters. 2. Nineteen models were evaluated to identify the best orders of Legendre polynomials. A model with Legendre polynomial of order 3 for additive genetic effect, order 3 for permanent environmental effects and order 1 for maternal permanent environmental effects was chosen as the best model. 3. According to the best model, phenotypic and genetic variances were higher at the end of the rearing period. Although direct heritability for BW reduced from 0.18 at hatch to 0.12 at 7 d of age, it gradually increased to 0.42 at 49 d of age. It indicates that BW at older ages is more controlled by genetic components in Japanese quail. 4. Phenotypic and genetic correlations between adjacent periods except hatching weight were more closely correlated than remote periods. The present results suggested that BW at earlier ages, especially at hatch, are different traits compared to BW at older ages. Therefore, BW at earlier ages could not be used as a selection criterion for improving BW at slaughter age.

Reference-free error estimation for multiple measurement methods.

PubMed

Madan, Hennadii; Pernuš, Franjo; Špiclin, Žiga

2018-01-01

We present a computational framework to select the most accurate and precise method of measurement of a certain quantity, when there is no access to the true value of the measurand. A typical use case is when several image analysis methods are applied to measure the value of a particular quantitative imaging biomarker from the same images. The accuracy of each measurement method is characterized by systematic error (bias), which is modeled as a polynomial in true values of measurand, and the precision as random error modeled with a Gaussian random variable. In contrast to previous works, the random errors are modeled jointly across all methods, thereby enabling the framework to analyze measurement methods based on similar principles, which may have correlated random errors. Furthermore, the posterior distribution of the error model parameters is estimated from samples obtained by Markov chain Monte-Carlo and analyzed to estimate the parameter values and the unknown true values of the measurand. The framework was validated on six synthetic and one clinical dataset containing measurements of total lesion load, a biomarker of neurodegenerative diseases, which was obtained with four automatic methods by analyzing brain magnetic resonance images. The estimates of bias and random error were in a good agreement with the corresponding least squares regression estimates against a reference.

Using Patient Demographics and Statistical Modeling to Predict Knee Tibia Component Sizing in Total Knee Arthroplasty.

PubMed

Ren, Anna N; Neher, Robert E; Bell, Tyler; Grimm, James

2018-06-01

Preoperative planning is important to achieve successful implantation in primary total knee arthroplasty (TKA). However, traditional TKA templating techniques are not accurate enough to predict the component size to a very close range. With the goal of developing a general predictive statistical model using patient demographic information, ordinal logistic regression was applied to build a proportional odds model to predict the tibia component size. The study retrospectively collected the data of 1992 primary Persona Knee System TKA procedures. Of them, 199 procedures were randomly selected as testing data and the rest of the data were randomly partitioned between model training data and model evaluation data with a ratio of 7:3. Different models were trained and evaluated on the training and validation data sets after data exploration. The final model had patient gender, age, weight, and height as independent variables and predicted the tibia size within 1 size difference 96% of the time on the validation data, 94% of the time on the testing data, and 92% on a prospective cadaver data set. The study results indicated the statistical model built by ordinal logistic regression can increase the accuracy of tibia sizing information for Persona Knee preoperative templating. This research shows statistical modeling may be used with radiographs to dramatically enhance the templating accuracy, efficiency, and quality. In general, this methodology can be applied to other TKA products when the data are applicable. Copyright © 2018 Elsevier Inc. All rights reserved.

Bayesian inference for multivariate meta-analysis Box-Cox transformation models for individual patient data with applications to evaluation of cholesterol lowering drugs

PubMed Central

Kim, Sungduk; Chen, Ming-Hui; Ibrahim, Joseph G.; Shah, Arvind K.; Lin, Jianxin

2013-01-01

In this paper, we propose a class of Box-Cox transformation regression models with multidimensional random effects for analyzing multivariate responses for individual patient data (IPD) in meta-analysis. Our modeling formulation uses a multivariate normal response meta-analysis model with multivariate random effects, in which each response is allowed to have its own Box-Cox transformation. Prior distributions are specified for the Box-Cox transformation parameters as well as the regression coefficients in this complex model, and the Deviance Information Criterion (DIC) is used to select the best transformation model. Since the model is quite complex, a novel Monte Carlo Markov chain (MCMC) sampling scheme is developed to sample from the joint posterior of the parameters. This model is motivated by a very rich dataset comprising 26 clinical trials involving cholesterol lowering drugs where the goal is to jointly model the three dimensional response consisting of Low Density Lipoprotein Cholesterol (LDL-C), High Density Lipoprotein Cholesterol (HDL-C), and Triglycerides (TG) (LDL-C, HDL-C, TG). Since the joint distribution of (LDL-C, HDL-C, TG) is not multivariate normal and in fact quite skewed, a Box-Cox transformation is needed to achieve normality. In the clinical literature, these three variables are usually analyzed univariately: however, a multivariate approach would be more appropriate since these variables are correlated with each other. A detailed analysis of these data is carried out using the proposed methodology. PMID:23580436

Bayesian inference for multivariate meta-analysis Box-Cox transformation models for individual patient data with applications to evaluation of cholesterol-lowering drugs.

PubMed

Kim, Sungduk; Chen, Ming-Hui; Ibrahim, Joseph G; Shah, Arvind K; Lin, Jianxin

2013-10-15

In this paper, we propose a class of Box-Cox transformation regression models with multidimensional random effects for analyzing multivariate responses for individual patient data in meta-analysis. Our modeling formulation uses a multivariate normal response meta-analysis model with multivariate random effects, in which each response is allowed to have its own Box-Cox transformation. Prior distributions are specified for the Box-Cox transformation parameters as well as the regression coefficients in this complex model, and the deviance information criterion is used to select the best transformation model. Because the model is quite complex, we develop a novel Monte Carlo Markov chain sampling scheme to sample from the joint posterior of the parameters. This model is motivated by a very rich dataset comprising 26 clinical trials involving cholesterol-lowering drugs where the goal is to jointly model the three-dimensional response consisting of low density lipoprotein cholesterol (LDL-C), high density lipoprotein cholesterol (HDL-C), and triglycerides (TG) (LDL-C, HDL-C, TG). Because the joint distribution of (LDL-C, HDL-C, TG) is not multivariate normal and in fact quite skewed, a Box-Cox transformation is needed to achieve normality. In the clinical literature, these three variables are usually analyzed univariately; however, a multivariate approach would be more appropriate because these variables are correlated with each other. We carry out a detailed analysis of these data by using the proposed methodology. Copyright © 2013 John Wiley & Sons, Ltd.

An empirical study using permutation-based resampling in meta-regression

PubMed Central

2012-01-01

Background In meta-regression, as the number of trials in the analyses decreases, the risk of false positives or false negatives increases. This is partly due to the assumption of normality that may not hold in small samples. Creation of a distribution from the observed trials using permutation methods to calculate P values may allow for less spurious findings. Permutation has not been empirically tested in meta-regression. The objective of this study was to perform an empirical investigation to explore the differences in results for meta-analyses on a small number of trials using standard large sample approaches verses permutation-based methods for meta-regression. Methods We isolated a sample of randomized controlled clinical trials (RCTs) for interventions that have a small number of trials (herbal medicine trials). Trials were then grouped by herbal species and condition and assessed for methodological quality using the Jadad scale, and data were extracted for each outcome. Finally, we performed meta-analyses on the primary outcome of each group of trials and meta-regression for methodological quality subgroups within each meta-analysis. We used large sample methods and permutation methods in our meta-regression modeling. We then compared final models and final P values between methods. Results We collected 110 trials across 5 intervention/outcome pairings and 5 to 10 trials per covariate. When applying large sample methods and permutation-based methods in our backwards stepwise regression the covariates in the final models were identical in all cases. The P values for the covariates in the final model were larger in 78% (7/9) of the cases for permutation and identical for 22% (2/9) of the cases. Conclusions We present empirical evidence that permutation-based resampling may not change final models when using backwards stepwise regression, but may increase P values in meta-regression of multiple covariates for relatively small amount of trials. PMID:22587815

Job strain and resting heart rate: a cross-sectional study in a Swedish random working sample.

PubMed

Eriksson, Peter; Schiöler, Linus; Söderberg, Mia; Rosengren, Annika; Torén, Kjell

2016-03-05

Numerous studies have reported an association between stressing work conditions and cardiovascular disease. However, more evidence is needed, and the etiological mechanisms are unknown. Elevated resting heart rate has emerged as a possible risk factor for cardiovascular disease, but little is known about the relation to work-related stress. This study therefore investigated the association between job strain, job control, and job demands and resting heart rate. We conducted a cross-sectional survey of randomly selected men and women in Västra Götalandsregionen, Sweden (West county of Sweden) (n = 1552). Information about job strain, job demands, job control, heart rate and covariates was collected during the period 2001-2004 as part of the INTERGENE/ADONIX research project. Six different linear regression models were used with adjustments for gender, age, BMI, smoking, education, and physical activity in the fully adjusted model. Job strain was operationalized as the log-transformed ratio of job demands over job control in the statistical analyses. No associations were seen between resting heart rate and job demands. Job strain was associated with elevated resting heart rate in the unadjusted model (linear regression coefficient 1.26, 95 % CI 0.14 to 2.38), but not in any of the extended models. Low job control was associated with elevated resting heart rate after adjustments for gender, age, BMI, and smoking (linear regression coefficient -0.18, 95 % CI -0.30 to -0.02). However, there were no significant associations in the fully adjusted model. Low job control and job strain, but not job demands, were associated with elevated resting heart rate. However, the observed associations were modest and may be explained by confounding effects.

Empirical Performance of Cross-Validation With Oracle Methods in a Genomics Context.

PubMed

Martinez, Josue G; Carroll, Raymond J; Müller, Samuel; Sampson, Joshua N; Chatterjee, Nilanjan

2011-11-01

When employing model selection methods with oracle properties such as the smoothly clipped absolute deviation (SCAD) and the Adaptive Lasso, it is typical to estimate the smoothing parameter by m-fold cross-validation, for example, m = 10. In problems where the true regression function is sparse and the signals large, such cross-validation typically works well. However, in regression modeling of genomic studies involving Single Nucleotide Polymorphisms (SNP), the true regression functions, while thought to be sparse, do not have large signals. We demonstrate empirically that in such problems, the number of selected variables using SCAD and the Adaptive Lasso, with 10-fold cross-validation, is a random variable that has considerable and surprising variation. Similar remarks apply to non-oracle methods such as the Lasso. Our study strongly questions the suitability of performing only a single run of m-fold cross-validation with any oracle method, and not just the SCAD and Adaptive Lasso.

Artificial bias typically neglected in comparisons of uncertain atmospheric data

NASA Astrophysics Data System (ADS)

Pitkänen, Mikko R. A.; Mikkonen, Santtu; Lehtinen, Kari E. J.; Lipponen, Antti; Arola, Antti

2016-09-01

Publications in atmospheric sciences typically neglect biases caused by regression dilution (bias of the ordinary least squares line fitting) and regression to the mean (RTM) in comparisons of uncertain data. We use synthetic observations mimicking real atmospheric data to demonstrate how the biases arise from random data uncertainties of measurements, model output, or satellite retrieval products. Further, we provide examples of typical methods of data comparisons that have a tendency to pronounce the biases. The results show, that data uncertainties can significantly bias data comparisons due to regression dilution and RTM, a fact that is known in statistics but disregarded in atmospheric sciences. Thus, we argue that often these biases are widely regarded as measurement or modeling errors, for instance, while they in fact are artificial. It is essential that atmospheric and geoscience communities become aware of and consider these features in research.

Angiogenic Signaling in Living Breast Tumor Models

DTIC Science & Technology

2007-06-01

Poisson distributed random noise is added in an amount relative to the desired signal to noise ratio. We fit the data using a regressive fitting...AD_________________ Award Number: W81XWH-05-1-0396 TITLE: Angiogenic Signaling in Living Breast...CONTRACT NUMBER Angiogenic Signaling in Living Breast Tumor Models 5b. GRANT NUMBER W81XWH-05-1-0396 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d

MIXOR: a computer program for mixed-effects ordinal regression analysis.

PubMed

Hedeker, D; Gibbons, R D

1996-03-01

MIXOR provides maximum marginal likelihood estimates for mixed-effects ordinal probit, logistic, and complementary log-log regression models. These models can be used for analysis of dichotomous and ordinal outcomes from either a clustered or longitudinal design. For clustered data, the mixed-effects model assumes that data within clusters are dependent. The degree of dependency is jointly estimated with the usual model parameters, thus adjusting for dependence resulting from clustering of the data. Similarly, for longitudinal data, the mixed-effects approach can allow for individual-varying intercepts and slopes across time, and can estimate the degree to which these time-related effects vary in the population of individuals. MIXOR uses marginal maximum likelihood estimation, utilizing a Fisher-scoring solution. For the scoring solution, the Cholesky factor of the random-effects variance-covariance matrix is estimated, along with the effects of model covariates. Examples illustrating usage and features of MIXOR are provided.

A mixed-effects regression model for longitudinal multivariate ordinal data.

PubMed

Liu, Li C; Hedeker, Donald

2006-03-01

A mixed-effects item response theory model that allows for three-level multivariate ordinal outcomes and accommodates multiple random subject effects is proposed for analysis of multivariate ordinal outcomes in longitudinal studies. This model allows for the estimation of different item factor loadings (item discrimination parameters) for the multiple outcomes. The covariates in the model do not have to follow the proportional odds assumption and can be at any level. Assuming either a probit or logistic response function, maximum marginal likelihood estimation is proposed utilizing multidimensional Gauss-Hermite quadrature for integration of the random effects. An iterative Fisher scoring solution, which provides standard errors for all model parameters, is used. An analysis of a longitudinal substance use data set, where four items of substance use behavior (cigarette use, alcohol use, marijuana use, and getting drunk or high) are repeatedly measured over time, is used to illustrate application of the proposed model.

Prediction of hourly PM2.5 using a space-time support vector regression model

NASA Astrophysics Data System (ADS)

Yang, Wentao; Deng, Min; Xu, Feng; Wang, Hang

2018-05-01

Real-time air quality prediction has been an active field of research in atmospheric environmental science. The existing methods of machine learning are widely used to predict pollutant concentrations because of their enhanced ability to handle complex non-linear relationships. However, because pollutant concentration data, as typical geospatial data, also exhibit spatial heterogeneity and spatial dependence, they may violate the assumptions of independent and identically distributed random variables in most of the machine learning methods. As a result, a space-time support vector regression model is proposed to predict hourly PM2.5 concentrations. First, to address spatial heterogeneity, spatial clustering is executed to divide the study area into several homogeneous or quasi-homogeneous subareas. To handle spatial dependence, a Gauss vector weight function is then developed to determine spatial autocorrelation variables as part of the input features. Finally, a local support vector regression model with spatial autocorrelation variables is established for each subarea. Experimental data on PM2.5 concentrations in Beijing are used to verify whether the results of the proposed model are superior to those of other methods.

Covariate Selection for Multilevel Models with Missing Data

PubMed Central

Marino, Miguel; Buxton, Orfeu M.; Li, Yi

2017-01-01

Missing covariate data hampers variable selection in multilevel regression settings. Current variable selection techniques for multiply-imputed data commonly address missingness in the predictors through list-wise deletion and stepwise-selection methods which are problematic. Moreover, most variable selection methods are developed for independent linear regression models and do not accommodate multilevel mixed effects regression models with incomplete covariate data. We develop a novel methodology that is able to perform covariate selection across multiply-imputed data for multilevel random effects models when missing data is present. Specifically, we propose to stack the multiply-imputed data sets from a multiple imputation procedure and to apply a group variable selection procedure through group lasso regularization to assess the overall impact of each predictor on the outcome across the imputed data sets. Simulations confirm the advantageous performance of the proposed method compared with the competing methods. We applied the method to reanalyze the Healthy Directions-Small Business cancer prevention study, which evaluated a behavioral intervention program targeting multiple risk-related behaviors in a working-class, multi-ethnic population. PMID:28239457

Time Advice and Learning Questions in Computer Simulations

ERIC Educational Resources Information Center

Rey, Gunter Daniel

2011-01-01

Students (N = 101) used an introductory text and a computer simulation to learn fundamental concepts about statistical analyses (e.g., analysis of variance, regression analysis and General Linear Model). Each learner was randomly assigned to one cell of a 2 (with or without time advice) x 3 (with learning questions and corrective feedback, with…

Digital Games, Design, and Learning: A Systematic Review and Meta-Analysis

ERIC Educational Resources Information Center

Clark, Douglas B.; Tanner-Smith, Emily E.; Killingsworth, Stephen S.

2016-01-01

In this meta-analysis, we systematically reviewed research on digital games and learning for K-16 students. We synthesized comparisons of game versus nongame conditions (i.e., media comparisons) and comparisons of augmented games versus standard game designs (i.e., value-added comparisons). We used random-effects meta-regression models with robust…

Least Principal Components Analysis (LPCA): An Alternative to Regression Analysis.

ERIC Educational Resources Information Center

Olson, Jeffery E.

Often, all of the variables in a model are latent, random, or subject to measurement error, or there is not an obvious dependent variable. When any of these conditions exist, an appropriate method for estimating the linear relationships among the variables is Least Principal Components Analysis. Least Principal Components are robust, consistent,…

Using Cluster Bootstrapping to Analyze Nested Data With a Few Clusters.

PubMed

Huang, Francis L

2018-04-01

Cluster randomized trials involving participants nested within intact treatment and control groups are commonly performed in various educational, psychological, and biomedical studies. However, recruiting and retaining intact groups present various practical, financial, and logistical challenges to evaluators and often, cluster randomized trials are performed with a low number of clusters (~20 groups). Although multilevel models are often used to analyze nested data, researchers may be concerned of potentially biased results due to having only a few groups under study. Cluster bootstrapping has been suggested as an alternative procedure when analyzing clustered data though it has seen very little use in educational and psychological studies. Using a Monte Carlo simulation that varied the number of clusters, average cluster size, and intraclass correlations, we compared standard errors using cluster bootstrapping with those derived using ordinary least squares regression and multilevel models. Results indicate that cluster bootstrapping, though more computationally demanding, can be used as an alternative procedure for the analysis of clustered data when treatment effects at the group level are of primary interest. Supplementary material showing how to perform cluster bootstrapped regressions using R is also provided.

ORACLE INEQUALITIES FOR THE LASSO IN THE COX MODEL

PubMed Central

Huang, Jian; Sun, Tingni; Ying, Zhiliang; Yu, Yi; Zhang, Cun-Hui

2013-01-01

We study the absolute penalized maximum partial likelihood estimator in sparse, high-dimensional Cox proportional hazards regression models where the number of time-dependent covariates can be larger than the sample size. We establish oracle inequalities based on natural extensions of the compatibility and cone invertibility factors of the Hessian matrix at the true regression coefficients. Similar results based on an extension of the restricted eigenvalue can be also proved by our method. However, the presented oracle inequalities are sharper since the compatibility and cone invertibility factors are always greater than the corresponding restricted eigenvalue. In the Cox regression model, the Hessian matrix is based on time-dependent covariates in censored risk sets, so that the compatibility and cone invertibility factors, and the restricted eigenvalue as well, are random variables even when they are evaluated for the Hessian at the true regression coefficients. Under mild conditions, we prove that these quantities are bounded from below by positive constants for time-dependent covariates, including cases where the number of covariates is of greater order than the sample size. Consequently, the compatibility and cone invertibility factors can be treated as positive constants in our oracle inequalities. PMID:24086091

ORACLE INEQUALITIES FOR THE LASSO IN THE COX MODEL.

PubMed

Huang, Jian; Sun, Tingni; Ying, Zhiliang; Yu, Yi; Zhang, Cun-Hui

2013-06-01

We study the absolute penalized maximum partial likelihood estimator in sparse, high-dimensional Cox proportional hazards regression models where the number of time-dependent covariates can be larger than the sample size. We establish oracle inequalities based on natural extensions of the compatibility and cone invertibility factors of the Hessian matrix at the true regression coefficients. Similar results based on an extension of the restricted eigenvalue can be also proved by our method. However, the presented oracle inequalities are sharper since the compatibility and cone invertibility factors are always greater than the corresponding restricted eigenvalue. In the Cox regression model, the Hessian matrix is based on time-dependent covariates in censored risk sets, so that the compatibility and cone invertibility factors, and the restricted eigenvalue as well, are random variables even when they are evaluated for the Hessian at the true regression coefficients. Under mild conditions, we prove that these quantities are bounded from below by positive constants for time-dependent covariates, including cases where the number of covariates is of greater order than the sample size. Consequently, the compatibility and cone invertibility factors can be treated as positive constants in our oracle inequalities.

Shrinkage Estimation of Varying Covariate Effects Based On Quantile Regression

PubMed Central

Peng, Limin; Xu, Jinfeng; Kutner, Nancy

2013-01-01

Varying covariate effects often manifest meaningful heterogeneity in covariate-response associations. In this paper, we adopt a quantile regression model that assumes linearity at a continuous range of quantile levels as a tool to explore such data dynamics. The consideration of potential non-constancy of covariate effects necessitates a new perspective for variable selection, which, under the assumed quantile regression model, is to retain variables that have effects on all quantiles of interest as well as those that influence only part of quantiles considered. Current work on l1-penalized quantile regression either does not concern varying covariate effects or may not produce consistent variable selection in the presence of covariates with partial effects, a practical scenario of interest. In this work, we propose a shrinkage approach by adopting a novel uniform adaptive LASSO penalty. The new approach enjoys easy implementation without requiring smoothing. Moreover, it can consistently identify the true model (uniformly across quantiles) and achieve the oracle estimation efficiency. We further extend the proposed shrinkage method to the case where responses are subject to random right censoring. Numerical studies confirm the theoretical results and support the utility of our proposals. PMID:25332515

Multitrait, Random Regression, or Simple Repeatability Model in High-Throughput Phenotyping Data Improve Genomic Prediction for Wheat Grain Yield.

PubMed

Sun, Jin; Rutkoski, Jessica E; Poland, Jesse A; Crossa, José; Jannink, Jean-Luc; Sorrells, Mark E

2017-07-01

High-throughput phenotyping (HTP) platforms can be used to measure traits that are genetically correlated with wheat ( L.) grain yield across time. Incorporating such secondary traits in the multivariate pedigree and genomic prediction models would be desirable to improve indirect selection for grain yield. In this study, we evaluated three statistical models, simple repeatability (SR), multitrait (MT), and random regression (RR), for the longitudinal data of secondary traits and compared the impact of the proposed models for secondary traits on their predictive abilities for grain yield. Grain yield and secondary traits, canopy temperature (CT) and normalized difference vegetation index (NDVI), were collected in five diverse environments for 557 wheat lines with available pedigree and genomic information. A two-stage analysis was applied for pedigree and genomic selection (GS). First, secondary traits were fitted by SR, MT, or RR models, separately, within each environment. Then, best linear unbiased predictions (BLUPs) of secondary traits from the above models were used in the multivariate prediction models to compare predictive abilities for grain yield. Predictive ability was substantially improved by 70%, on average, from multivariate pedigree and genomic models when including secondary traits in both training and test populations. Additionally, (i) predictive abilities slightly varied for MT, RR, or SR models in this data set, (ii) results indicated that including BLUPs of secondary traits from the MT model was the best in severe drought, and (iii) the RR model was slightly better than SR and MT models under drought environment. Copyright © 2017 Crop Science Society of America.

Gender conformity, self-objectification, and body image for sorority and nonsorority women: A closer look.

PubMed

Adams, David Francis; Behrens, Erica; Gann, Lianne; Schoen, Eva

2017-01-01

Sororities have been identified as placing young women at risk for body image concerns due to a focus on traditional gender role norms and objectification of women. This study assessed the relationship between conformity to feminine gender role norms, self-objectification, and body image surveillance among undergraduate women. In a random sample of undergraduates, the authors examined data from sorority and nonsorority women. In a random sample of undergraduate women, the authors assessed the impact of traditional feminine gender role norms on self-objectification, body image, and feedback regarding physical appearance for sorority and nonsorority undergraduate women. Three linear regressions were conducted, and only conformity to feminine gender role norms contributed significantly in each regression model. Regardless of sorority membership, conformity to feminine gender role norms was found to significantly contribute to increased body consciousness, negative body image, and feedback on physical appearance.

Easy and low-cost identification of metabolic syndrome in patients treated with second-generation antipsychotics: artificial neural network and logistic regression models.

PubMed

Lin, Chao-Cheng; Bai, Ya-Mei; Chen, Jen-Yeu; Hwang, Tzung-Jeng; Chen, Tzu-Ting; Chiu, Hung-Wen; Li, Yu-Chuan

2010-03-01

Metabolic syndrome (MetS) is an important side effect of second-generation antipsychotics (SGAs). However, many SGA-treated patients with MetS remain undetected. In this study, we trained and validated artificial neural network (ANN) and multiple logistic regression models without biochemical parameters to rapidly identify MetS in patients with SGA treatment. A total of 383 patients with a diagnosis of schizophrenia or schizoaffective disorder (DSM-IV criteria) with SGA treatment for more than 6 months were investigated to determine whether they met the MetS criteria according to the International Diabetes Federation. The data for these patients were collected between March 2005 and September 2005. The input variables of ANN and logistic regression were limited to demographic and anthropometric data only. All models were trained by randomly selecting two-thirds of the patient data and were internally validated with the remaining one-third of the data. The models were then externally validated with data from 69 patients from another hospital, collected between March 2008 and June 2008. The area under the receiver operating characteristic curve (AUC) was used to measure the performance of all models. Both the final ANN and logistic regression models had high accuracy (88.3% vs 83.6%), sensitivity (93.1% vs 86.2%), and specificity (86.9% vs 83.8%) to identify MetS in the internal validation set. The mean +/- SD AUC was high for both the ANN and logistic regression models (0.934 +/- 0.033 vs 0.922 +/- 0.035, P = .63). During external validation, high AUC was still obtained for both models. Waist circumference and diastolic blood pressure were the common variables that were left in the final ANN and logistic regression models. Our study developed accurate ANN and logistic regression models to detect MetS in patients with SGA treatment. The models are likely to provide a noninvasive tool for large-scale screening of MetS in this group of patients. (c) 2010 Physicians Postgraduate Press, Inc.

The steroids for corneal ulcers trial (SCUT): secondary 12-month clinical outcomes of a randomized controlled trial.

PubMed

Srinivasan, Muthiah; Mascarenhas, Jeena; Rajaraman, Revathi; Ravindran, Meenakshi; Lalitha, Prajna; O'Brien, Kieran S; Glidden, David V; Ray, Kathryn J; Oldenburg, Catherine E; Zegans, Michael E; Whitcher, John P; McLeod, Stephen D; Porco, Travis C; Lietman, Thomas M; Acharya, Nisha R

2014-02-01

To determine whether topical corticosteroids as adjunctive therapy for bacterial keratitis improves long-term clinical outcomes. Randomized, placebo-controlled, double-masked clinical trial. This multicenter trial compared 1.0% prednisolone sodium phosphate to placebo in the treatment of bacterial keratitis among 500 patients with culture-positive ulcers receiving 48 hours of moxifloxacin before randomization. The primary endpoint was 3 months from enrollment, and 399 patients were evaluated at 12 months. The outcomes examined were best spectacle-corrected visual acuity (BSCVA) and scar size at 12 months. Based on previous results, regression models with adjustments for baseline status and/or causative organism were used for analysis. No significant differences in clinical outcomes by treatment group were seen with the prespecified regression models (BSCVA: -0.04 logMAR, 95% CI, -0.12 to 0.05, P = .39; scar size: 0.03 mm, 95% CI, -0.12 to 0.18, P = .69). A regression model including a Nocardia-treatment arm interaction found corticosteroid use associated with a mean 1-line improvement in BSCVA at 12 months among patients with non-Nocardia ulcers (-0.10 logMAR, 95% CI, -0.19 to -0.02, P = .02). No significant difference was observed in 12-month BSCVA for Nocardia ulcers (0.18 logMAR, 95% CI, -0.04 to 0.41, P = .16). Corticosteroids were associated with larger mean scar size at 12 months among Nocardia ulcers (0.47 mm, 95% CI, 0.06-0.88, P = .02) and no significant difference was identified by treatment for scar size for non-Nocardia ulcers (-0.06 mm, 95% CI, -0.21 to 0.10, P = .46). Adjunctive topical corticosteroid therapy may be associated with improved long-term clinical outcomes in bacterial corneal ulcers not caused by Nocardia species. Copyright © 2014 Elsevier Inc. All rights reserved.

The Steroids for Corneal Ulcers Trial (SCUT): secondary 12-month clinical outcomes of a randomized controlled trial

PubMed Central

Srinivasan, Muthiah; Mascarenhas, Jeena; Rajaraman, Revathi; Ravindran, Meenakshi; Lalitha, Prajna; O’Brien, Kieran S.; Glidden, David V.; Ray, Kathryn J.; Oldenburg, Catherine E.; Zegans, Michael E.; Whitcher, John P.; McLeod, Stephen D.; Porco, Travis C.; Lietman, Thomas M.; Acharya, Nisha R.

2013-01-01

Purpose To determine whether topical corticosteroids as adjunctive therapy for bacterial keratitis improves long-term clinical outcomes. Design Randomized placebo-controlled double-masked clinical trial. Methods This multicenter trial compared 1.0% prednisolone sodium phosphate to placebo in the treatment of bacterial keratitis among 500 patients with culture-positive ulcers receiving 48 hours of moxifloxacin before randomization. The primary endpoint was 3 months from enrollment, and 399 patients were evaluated at 12 months. The outcomes examined were best spectacle-corrected visual acuity (BSCVA) and scar size at 12 months. Based on previous results, regression models with adjustments for baseline status and/or causative organism were used for analysis. Results No significant differences in clinical outcomes by treatment group were seen with the pre-specified regression models (BSCVA: −0.04 logMAR, 95%CI, −0.12 to 0.05, P=0.39; scar size: 0.03mm, 95% CI, −0.12 to 0.18, P=0.69). A regression model including a Nocardia-treatment arm interaction found corticosteroid use associated with a mean one-line improvement in BSCVA at 12 months among patients with non-Nocardia ulcers (−0.10 logMAR, 95%CI, −0.19 to −0.02, P=0.02). No significant difference was observed in 12-month BSCVA for Nocardia ulcers (0.18 logMAR, 95% CI, −0.04 to 0.41, P=0.16). Corticosteroids were associated with larger mean scar size at 12 months among Nocardia ulcers (0.47mm, 95% CI, 0.06 to 0.88, P=0.02) and no significant difference was identified by treatment for scar size for non-Nocardia ulcers (−0.06mm, 95%CI, −0.21 to 0.10, P=0.46). Conclusions Adjunctive topical corticosteroid therapy may be associated with improved long-term clinical outcomes in bacterial corneal ulcers not caused by Nocardia species. PMID:24315294

The comparison of landslide ratio-based and general logistic regression landslide susceptibility models in the Chishan watershed after 2009 Typhoon Morakot

NASA Astrophysics Data System (ADS)

WU, Chunhung

2015-04-01

The research built the original logistic regression landslide susceptibility model (abbreviated as or-LRLSM) and landslide ratio-based ogistic regression landslide susceptibility model (abbreviated as lr-LRLSM), compared the performance and explained the error source of two models. The research assumes that the performance of the logistic regression model can be better if the distribution of landslide ratio and weighted value of each variable is similar. Landslide ratio is the ratio of landslide area to total area in the specific area and an useful index to evaluate the seriousness of landslide disaster in Taiwan. The research adopted the landside inventory induced by 2009 Typhoon Morakot in the Chishan watershed, which was the most serious disaster event in the last decade, in Taiwan. The research adopted the 20 m grid as the basic unit in building the LRLSM, and six variables, including elevation, slope, aspect, geological formation, accumulated rainfall, and bank erosion, were included in the two models. The six variables were divided as continuous variables, including elevation, slope, and accumulated rainfall, and categorical variables, including aspect, geological formation and bank erosion in building the or-LRLSM, while all variables, which were classified based on landslide ratio, were categorical variables in building the lr-LRLSM. Because the count of whole basic unit in the Chishan watershed was too much to calculate by using commercial software, the research took random sampling instead of the whole basic units. The research adopted equal proportions of landslide unit and not landslide unit in logistic regression analysis. The research took 10 times random sampling and selected the group with the best Cox & Snell R2 value and Nagelkerker R2 value as the database for the following analysis. Based on the best result from 10 random sampling groups, the or-LRLSM (lr-LRLSM) is significant at the 1% level with Cox & Snell R2 = 0.190 (0.196) and Nagelkerke R2 = 0.253 (0.260). The unit with the landslide susceptibility value > 0.5 (≦ 0.5) will be classified as a predicted landslide unit (not landslide unit). The AUC, i.e. the area under the relative operating characteristic curve, of or-LRLSM in the Chishan watershed is 0.72, while that of lr-LRLSM is 0.77. Furthermore, the average correct ratio of lr-LRLSM (73.3%) is better than that of or-LRLSM (68.3%). The research analyzed in detail the error sources from the two models. In continuous variables, using the landslide ratio-based classification in building the lr-LRLSM can let the distribution of weighted value more similar to distribution of landslide ratio in the range of continuous variable than that in building the or-LRLSM. In categorical variables, the meaning of using the landslide ratio-based classification in building the lr-LRLSM is to gather the parameters with approximate landslide ratio together. The mean correct ratio in continuous variables (categorical variables) by using the lr-LRLSM is better than that in or-LRLSM by 0.6 ~ 2.6% (1.7% ~ 6.0%). Building the landslide susceptibility model by using landslide ratio-based classification is practical and of better performance than that by using the original logistic regression.

Modeling lactation curves and estimation of genetic parameters in Holstein cows using multiple-trait random regression models.

PubMed

Kheirabadi, Khabat; Rashidi, Amir; Alijani, Sadegh; Imumorin, Ikhide

2014-11-01

We compared the goodness of fit of three mathematical functions (including: Legendre polynomials, Lidauer-Mäntysaari function and Wilmink function) for describing the lactation curve of primiparous Iranian Holstein cows by using multiple-trait random regression models (MT-RRM). Lactational submodels provided the largest daily additive genetic (AG) and permanent environmental (PE) variance estimates at the end and at the onset of lactation, respectively, as well as low genetic correlations between peripheral test-day records. For all models, heritability estimates were highest at the end of lactation (245 to 305 days) and ranged from 0.05 to 0.26, 0.03 to 0.12 and 0.04 to 0.24 for milk, fat and protein yields, respectively. Generally, the genetic correlations between traits depend on how far apart they are or whether they are on the same day in any two traits. On average, genetic correlations between milk and fat were the lowest and those between fat and protein were intermediate, while those between milk and protein were the highest. Results from all criteria (Akaike's and Schwarz's Bayesian information criterion, and -2*logarithm of the likelihood function) suggested that a model with 2 and 5 coefficients of Legendre polynomials for AG and PE effects, respectively, was the most adequate for fitting the data. © 2014 Japanese Society of Animal Science.

Random regression analysis for body weights and main morphological traits in genetically improved farmed tilapia (Oreochromis niloticus).

PubMed

He, Jie; Zhao, Yunfeng; Zhao, Jingli; Gao, Jin; Xu, Pao; Yang, Runqing

2018-02-01

To genetically analyse growth traits in genetically improved farmed tilapia (GIFT), the body weight (BWE) and main morphological traits, including body length (BL), body depth (BD), body width (BWI), head length (HL) and length of the caudal peduncle (CPL), were measured six times in growth duration on 1451 fish from 45 mixed families of full and half sibs. A random regression model (RRM) was used to model genetic changes of the growth traits with days of age and estimate the heritability for any growth point and genetic correlations between pairwise growth points. Using the covariance function based on optimal RRMs, the heritabilities were estimated to be from 0.102 to 0.662 for BWE, 0.157 to 0.591 for BL, 0.047 to 0.621 for BD, 0.018 to 0.577 for BWI, 0.075 to 0.597 for HL and 0.032 to 0.610 for CPL between 60 and 140 days of age. All genetic correlations exceeded 0.5 between pairwise growth points. Moreover, the traits at initial days of age showed less correlation with those at later days of age. With phenotypes observed repeatedly, the model choice showed that the optimal RRMs could more precisely predict breeding values at a specific growth time than repeatability models or multiple trait animal models, which enhanced the efficiency of selection for the BWE and main morphological traits.

Examining the influence of link function misspecification in conventional regression models for developing crash modification factors.

PubMed

Wu, Lingtao; Lord, Dominique

2017-05-01

This study further examined the use of regression models for developing crash modification factors (CMFs), specifically focusing on the misspecification in the link function. The primary objectives were to validate the accuracy of CMFs derived from the commonly used regression models (i.e., generalized linear models or GLMs with additive linear link functions) when some of the variables have nonlinear relationships and quantify the amount of bias as a function of the nonlinearity. Using the concept of artificial realistic data, various linear and nonlinear crash modification functions (CM-Functions) were assumed for three variables. Crash counts were randomly generated based on these CM-Functions. CMFs were then derived from regression models for three different scenarios. The results were compared with the assumed true values. The main findings are summarized as follows: (1) when some variables have nonlinear relationships with crash risk, the CMFs for these variables derived from the commonly used GLMs are all biased, especially around areas away from the baseline conditions (e.g., boundary areas); (2) with the increase in nonlinearity (i.e., nonlinear relationship becomes stronger), the bias becomes more significant; (3) the quality of CMFs for other variables having linear relationships can be influenced when mixed with those having nonlinear relationships, but the accuracy may still be acceptable; and (4) the misuse of the link function for one or more variables can also lead to biased estimates for other parameters. This study raised the importance of the link function when using regression models for developing CMFs. Copyright © 2017 Elsevier Ltd. All rights reserved.

Regression calibration for models with two predictor variables measured with error and their interaction, using instrumental variables and longitudinal data.

PubMed

Strand, Matthew; Sillau, Stefan; Grunwald, Gary K; Rabinovitch, Nathan

2014-02-10

Regression calibration provides a way to obtain unbiased estimators of fixed effects in regression models when one or more predictors are measured with error. Recent development of measurement error methods has focused on models that include interaction terms between measured-with-error predictors, and separately, methods for estimation in models that account for correlated data. In this work, we derive explicit and novel forms of regression calibration estimators and associated asymptotic variances for longitudinal models that include interaction terms, when data from instrumental and unbiased surrogate variables are available but not the actual predictors of interest. The longitudinal data are fit using linear mixed models that contain random intercepts and account for serial correlation and unequally spaced observations. The motivating application involves a longitudinal study of exposure to two pollutants (predictors) - outdoor fine particulate matter and cigarette smoke - and their association in interactive form with levels of a biomarker of inflammation, leukotriene E4 (LTE 4 , outcome) in asthmatic children. Because the exposure concentrations could not be directly observed, we used measurements from a fixed outdoor monitor and urinary cotinine concentrations as instrumental variables, and we used concentrations of fine ambient particulate matter and cigarette smoke measured with error by personal monitors as unbiased surrogate variables. We applied the derived regression calibration methods to estimate coefficients of the unobserved predictors and their interaction, allowing for direct comparison of toxicity of the different pollutants. We used simulations to verify accuracy of inferential methods based on asymptotic theory. Copyright © 2013 John Wiley & Sons, Ltd.

Comparison of partial least squares and random forests for evaluating relationship between phenolics and bioactivities of Neptunia oleracea.

PubMed

Lee, Soo Yee; Mediani, Ahmed; Maulidiani, Maulidiani; Khatib, Alfi; Ismail, Intan Safinar; Zawawi, Norhasnida; Abas, Faridah

2018-01-01

Neptunia oleracea is a plant consumed as a vegetable and which has been used as a folk remedy for several diseases. Herein, two regression models (partial least squares, PLS; and random forest, RF) in a metabolomics approach were compared and applied to the evaluation of the relationship between phenolics and bioactivities of N. oleracea. In addition, the effects of different extraction conditions on the phenolic constituents were assessed by pattern recognition analysis. Comparison of the PLS and RF showed that RF exhibited poorer generalization and hence poorer predictive performance. Both the regression coefficient of PLS and the variable importance of RF revealed that quercetin and kaempferol derivatives, caffeic acid and vitexin-2-O-rhamnoside were significant towards the tested bioactivities. Furthermore, principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA) results showed that sonication and absolute ethanol are the preferable extraction method and ethanol ratio, respectively, to produce N. oleracea extracts with high phenolic levels and therefore high DPPH scavenging and α-glucosidase inhibitory activities. Both PLS and RF are useful regression models in metabolomics studies. This work provides insight into the performance of different multivariate data analysis tools and the effects of different extraction conditions on the extraction of desired phenolics from plants. © 2017 Society of Chemical Industry. © 2017 Society of Chemical Industry.

Modeling Longitudinal Data Containing Non-Normal Within Subject Errors

NASA Technical Reports Server (NTRS)

Feiveson, Alan; Glenn, Nancy L.

2013-01-01

The mission of the National Aeronautics and Space Administration’s (NASA) human research program is to advance safe human spaceflight. This involves conducting experiments, collecting data, and analyzing data. The data are longitudinal and result from a relatively few number of subjects; typically 10 – 20. A longitudinal study refers to an investigation where participant outcomes and possibly treatments are collected at multiple follow-up times. Standard statistical designs such as mean regression with random effects and mixed–effects regression are inadequate for such data because the population is typically not approximately normally distributed. Hence, more advanced data analysis methods are necessary. This research focuses on four such methods for longitudinal data analysis: the recently proposed linear quantile mixed models (lqmm) by Geraci and Bottai (2013), quantile regression, multilevel mixed–effects linear regression, and robust regression. This research also provides computational algorithms for longitudinal data that scientists can directly use for human spaceflight and other longitudinal data applications, then presents statistical evidence that verifies which method is best for specific situations. This advances the study of longitudinal data in a broad range of applications including applications in the sciences, technology, engineering and mathematics fields.

Bayesian Regression with Network Prior: Optimal Bayesian Filtering Perspective

PubMed Central

Qian, Xiaoning; Dougherty, Edward R.

2017-01-01

The recently introduced intrinsically Bayesian robust filter (IBRF) provides fully optimal filtering relative to a prior distribution over an uncertainty class ofjoint random process models, whereas formerly the theory was limited to model-constrained Bayesian robust filters, for which optimization was limited to the filters that are optimal for models in the uncertainty class. This paper extends the IBRF theory to the situation where there are both a prior on the uncertainty class and sample data. The result is optimal Bayesian filtering (OBF), where optimality is relative to the posterior distribution derived from the prior and the data. The IBRF theories for effective characteristics and canonical expansions extend to the OBF setting. A salient focus of the present work is to demonstrate the advantages of Bayesian regression within the OBF setting over the classical Bayesian approach in the context otlinear Gaussian models. PMID:28824268

Mapping Regional Impervious Surface Distribution from Night Time Light: The Variability across Global Cities

NASA Astrophysics Data System (ADS)

Lin, M.; Yang, Z.; Park, H.; Qian, S.; Chen, J.; Fan, P.

2017-12-01

Impervious surface area (ISA) has become an important indicator for studying urban environments, but mapping ISA at the regional or global scale is still challenging due to the complexity of impervious surface features. The Defense Meteorological Satellite Program's Operational Linescan System (DMSP-OLS) nighttime light data is (NTL) and Resolution Imaging Spectroradiometer (MODIS) are the major remote sensing data source for regional ISA mapping. A single regression relationship between fractional ISA and NTL or various index derived based on NTL and MODIS vegetation index (NDVI) data was established in many previous studies for regional ISA mapping. However, due to the varying geographical, climatic, and socio-economic characteristics of different cities, the same regression relationship may vary significantly across different cities in the same region in terms of both fitting performance (i.e. R2) and the rate of change (Slope). In this study, we examined the regression relationship between fractional ISA and Vegetation Adjusted Nighttime light Urban Index (VANUI) for 120 randomly selected cities around the world with a multilevel regression model. We found that indeed there is substantial variability of both the R2 (0.68±0.29) and slopes (0.64±0.40) among individual regressions, which suggests that multilevel/hierarchical models are needed for accuracy improvement of future regional ISA mapping .Further analysis also let us find the this substantial variability are affected by climate conditions, socio-economic status, and urban spatial structures. However, all these effects are nonlinear rather than linear, thus could not modeled explicitly in multilevel linear regression models.

Concepts for a theoretical and experimental study of lifting rotor random loads and vibrations. Phase 6-B: Experiments with progressing/regressing forced rotor flapping modes

NASA Technical Reports Server (NTRS)

Hohenemser, K. H.; Crews, S. T.

1972-01-01

A two bladed 16-inch hingeless rotor model was built and tested outside and inside a 24 by 24 inch wind tunnel test section at collective pitch settings up to 5 deg and rotor advance ratios up to .4. The rotor model has a simple eccentric mechanism to provide progressing or regressing cyclic pitch excitation. The flapping responses were compared to analytically determined responses which included flap-bending elasticity but excluded rotor wake effects. Substantial systematic deviations of the measured responses from the computed responses were found, which were interpreted as the effects of interaction of the blades with a rotating asymmetrical wake.

Concentration and flux of total and dissolved phosphorus, total nitrogen, chloride, and total suspended solids for monitored tributaries of Lake Champlain, 1990-2012

USGS Publications Warehouse

Medalie, Laura

2014-01-01

Annual and daily concentrations and fluxes of total and dissolved phosphorus, total nitrogen, chloride, and total suspended solids were estimated for 18 monitored tributaries to Lake Champlain by using the Weighted Regressions on Time, Discharge, and Seasons regression model. Estimates were made for 21 or 23 years, depending on data availability, for the purpose of providing timely and accessible summary reports as stipulated in the 2010 update to the Lake Champlain “Opportunities for Action” management plan. Estimates of concentration and flux were provided for each tributary based on (1) observed daily discharges and (2) a flow-normalizing procedure, which removed the random fluctuations of climate-related variability. The flux bias statistic, an indicator of the ability of the Weighted Regressions on Time, Discharge, and Season regression models to provide accurate representations of flux, showed acceptable bias (less than ±10 percent) for 68 out of 72 models for total and dissolved phosphorus, total nitrogen, and chloride. Six out of 18 models for total suspended solids had moderate bias (between 10 and 30 percent), an expected result given the frequently nonlinear relation between total suspended solids and discharge. One model for total suspended solids with a very high bias was influenced by a single extreme value; however, removal of that value, although reducing the bias substantially, had little effect on annual fluxes.

Can adverse maternal and perinatal outcomes be predicted when blood pressure becomes elevated? Secondary analyses from the CHIPS (Control of Hypertension In Pregnancy Study) randomized controlled trial.

PubMed

Magee, Laura A; von Dadelszen, Peter; Singer, Joel; Lee, Terry; Rey, Evelyne; Ross, Susan; Asztalos, Elizabeth; Murphy, Kellie E; Menzies, Jennifer; Sanchez, Johanna; Gafni, Amiram; Gruslin, Andrée; Helewa, Michael; Hutton, Eileen; Lee, Shoo K; Logan, Alexander G; Ganzevoort, Wessel; Welch, Ross; Thornton, Jim G; Moutquin, Jean Marie

2016-07-01

For women with chronic or gestational hypertension in CHIPS (Control of Hypertension In Pregnancy Study, NCT01192412), we aimed to examine whether clinical predictors collected at randomization could predict adverse outcomes. This was a planned, secondary analysis of data from the 987 women in the CHIPS Trial. Logistic regression was used to examine the impact of 19 candidate predictors on the probability of adverse perinatal (pregnancy loss or high level neonatal care for >48 h, or birthweight <10th percentile) or maternal outcomes (severe hypertension, preeclampsia, or delivery at <34 or <37 weeks). A model containing all candidate predictors was used to start the stepwise regression process based on goodness of fit as measured by the Akaike information criterion. For face validity, these variables were forced into the model: treatment group ("less tight" or "tight" control), antihypertensive type at randomization, and blood pressure within 1 week before randomization. Continuous variables were represented continuously or dichotomized based on the smaller p-value in univariate analyses. An area-under-the-receiver-operating-curve (AUC ROC) of ≥0.70 was taken to reflect a potentially useful model. Point estimates for AUC ROC were <0.70 for all but severe hypertension (0.70, 95% CI 0.67-0.74) and delivery at <34 weeks (0.71, 95% CI 0.66-0.75). Therefore, no model warranted further assessment of performance. CHIPS data suggest that when women with chronic hypertension develop an elevated blood pressure in pregnancy, or formerly normotensive women develop new gestational hypertension, maternal and current pregnancy clinical characteristics cannot predict adverse outcomes in the index pregnancy. © 2016 The Authors. Acta Obstetricia et Gynecologica Scandinavica published by John Wiley & Sons Ltd on behalf of Nordic Federation of Societies of Obstetrics and Gynecology (NFOG).

Predicting Pre-planting Risk of Stagonospora nodorum blotch in Winter Wheat Using Machine Learning Models.

PubMed

Mehra, Lucky K; Cowger, Christina; Gross, Kevin; Ojiambo, Peter S

2016-01-01

Pre-planting factors have been associated with the late-season severity of Stagonospora nodorum blotch (SNB), caused by the fungal pathogen Parastagonospora nodorum, in winter wheat (Triticum aestivum). The relative importance of these factors in the risk of SNB has not been determined and this knowledge can facilitate disease management decisions prior to planting of the wheat crop. In this study, we examined the performance of multiple regression (MR) and three machine learning algorithms namely artificial neural networks, categorical and regression trees, and random forests (RF), in predicting the pre-planting risk of SNB in wheat. Pre-planting factors tested as potential predictor variables were cultivar resistance, latitude, longitude, previous crop, seeding rate, seed treatment, tillage type, and wheat residue. Disease severity assessed at the end of the growing season was used as the response variable. The models were developed using 431 disease cases (unique combinations of predictors) collected from 2012 to 2014 and these cases were randomly divided into training, validation, and test datasets. Models were evaluated based on the regression of observed against predicted severity values of SNB, sensitivity-specificity ROC analysis, and the Kappa statistic. A strong relationship was observed between late-season severity of SNB and specific pre-planting factors in which latitude, longitude, wheat residue, and cultivar resistance were the most important predictors. The MR model explained 33% of variability in the data, while machine learning models explained 47 to 79% of the total variability. Similarly, the MR model correctly classified 74% of the disease cases, while machine learning models correctly classified 81 to 83% of these cases. Results show that the RF algorithm, which explained 79% of the variability within the data, was the most accurate in predicting the risk of SNB, with an accuracy rate of 93%. The RF algorithm could allow early assessment of the risk of SNB, facilitating sound disease management decisions prior to planting of wheat.

Bayesian Nonparametric Inference – Why and How

PubMed Central

Müller, Peter; Mitra, Riten

2013-01-01

We review inference under models with nonparametric Bayesian (BNP) priors. The discussion follows a set of examples for some common inference problems. The examples are chosen to highlight problems that are challenging for standard parametric inference. We discuss inference for density estimation, clustering, regression and for mixed effects models with random effects distributions. While we focus on arguing for the need for the flexibility of BNP models, we also review some of the more commonly used BNP models, thus hopefully answering a bit of both questions, why and how to use BNP. PMID:24368932

Assessment of wastewater treatment facility compliance with decreasing ammonia discharge limits using a regression tree model.

PubMed

Suchetana, Bihu; Rajagopalan, Balaji; Silverstein, JoAnn

2017-11-15

A regression tree-based diagnostic approach is developed to evaluate factors affecting US wastewater treatment plant compliance with ammonia discharge permit limits using Discharge Monthly Report (DMR) data from a sample of 106 municipal treatment plants for the period of 2004-2008. Predictor variables used to fit the regression tree are selected using random forests, and consist of the previous month's effluent ammonia, influent flow rates and plant capacity utilization. The tree models are first used to evaluate compliance with existing ammonia discharge standards at each facility and then applied assuming more stringent discharge limits, under consideration in many states. The model predicts that the ability to meet both current and future limits depends primarily on the previous month's treatment performance. With more stringent discharge limits predicted ammonia concentration relative to the discharge limit, increases. In-sample validation shows that the regression trees can provide a median classification accuracy of >70%. The regression tree model is validated using ammonia discharge data from an operating wastewater treatment plant and is able to accurately predict the observed ammonia discharge category approximately 80% of the time, indicating that the regression tree model can be applied to predict compliance for individual treatment plants providing practical guidance for utilities and regulators with an interest in controlling ammonia discharges. The proposed methodology is also used to demonstrate how to delineate reliable sources of demand and supply in a point source-to-point source nutrient credit trading scheme, as well as how planners and decision makers can set reasonable discharge limits in future. Copyright © 2017 Elsevier B.V. All rights reserved.

Comparing spatial regression to random forests for large environmental data sets

EPA Science Inventory

Environmental data may be “large” due to number of records, number of covariates, or both. Random forests has a reputation for good predictive performance when using many covariates, whereas spatial regression, when using reduced rank methods, has a reputatio...

Evaluating an Organizational-Level Occupational Health Intervention in a Combined Regression Discontinuity and Randomized Control Design.

PubMed

Sørensen, By Ole H

2016-10-01

Organizational-level occupational health interventions have great potential to improve employees' health and well-being. However, they often compare unfavourably to individual-level interventions. This calls for improving methods for designing, implementing and evaluating organizational interventions. This paper presents and discusses the regression discontinuity design because, like the randomized control trial, it is a strong summative experimental design, but it typically fits organizational-level interventions better. The paper explores advantages and disadvantages of a regression discontinuity design with an embedded randomized control trial. It provides an example from an intervention study focusing on reducing sickness absence in 196 preschools. The paper demonstrates that such a design fits the organizational context, because it allows management to focus on organizations or workgroups with the most salient problems. In addition, organizations may accept an embedded randomized design because the organizations or groups with most salient needs receive obligatory treatment as part of the regression discontinuity design. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

Random sample consensus combined with partial least squares regression (RANSAC-PLS) for microbial metabolomics data mining and phenotype improvement.

PubMed

Teoh, Shao Thing; Kitamura, Miki; Nakayama, Yasumune; Putri, Sastia; Mukai, Yukio; Fukusaki, Eiichiro

2016-08-01

In recent years, the advent of high-throughput omics technology has made possible a new class of strain engineering approaches, based on identification of possible gene targets for phenotype improvement from omic-level comparison of different strains or growth conditions. Metabolomics, with its focus on the omic level closest to the phenotype, lends itself naturally to this semi-rational methodology. When a quantitative phenotype such as growth rate under stress is considered, regression modeling using multivariate techniques such as partial least squares (PLS) is often used to identify metabolites correlated with the target phenotype. However, linear modeling techniques such as PLS require a consistent metabolite-phenotype trend across the samples, which may not be the case when outliers or multiple conflicting trends are present in the data. To address this, we proposed a data-mining strategy that utilizes random sample consensus (RANSAC) to select subsets of samples with consistent trends for construction of better regression models. By applying a combination of RANSAC and PLS (RANSAC-PLS) to a dataset from a previous study (gas chromatography/mass spectrometry metabolomics data and 1-butanol tolerance of 19 yeast mutant strains), new metabolites were indicated to be correlated with tolerance within certain subsets of the samples. The relevance of these metabolites to 1-butanol tolerance were then validated from single-deletion strains of corresponding metabolic genes. The results showed that RANSAC-PLS is a promising strategy to identify unique metabolites that provide additional hints for phenotype improvement, which could not be detected by traditional PLS modeling using the entire dataset. Copyright © 2016 The Society for Biotechnology, Japan. Published by Elsevier B.V. All rights reserved.

An overview of longitudinal data analysis methods for neurological research.

PubMed

Locascio, Joseph J; Atri, Alireza

2011-01-01

The purpose of this article is to provide a concise, broad and readily accessible overview of longitudinal data analysis methods, aimed to be a practical guide for clinical investigators in neurology. In general, we advise that older, traditional methods, including (1) simple regression of the dependent variable on a time measure, (2) analyzing a single summary subject level number that indexes changes for each subject and (3) a general linear model approach with a fixed-subject effect, should be reserved for quick, simple or preliminary analyses. We advocate the general use of mixed-random and fixed-effect regression models for analyses of most longitudinal clinical studies. Under restrictive situations or to provide validation, we recommend: (1) repeated-measure analysis of covariance (ANCOVA), (2) ANCOVA for two time points, (3) generalized estimating equations and (4) latent growth curve/structural equation models.

The role of gender in a smoking cessation intervention: a cluster randomized clinical trial

PubMed Central

2011-01-01

Background The prevalence of smoking in Spain is high in both men and women. The aim of our study was to evaluate the role of gender in the effectiveness of a specific smoking cessation intervention conducted in Spain. Methods This study was a secondary analysis of a cluster randomized clinical trial in which the randomization unit was the Basic Care Unit (family physician and nurse who care for the same group of patients). The intervention consisted of a six-month period of implementing the recommendations of a Clinical Practice Guideline. A total of 2,937 current smokers at 82 Primary Care Centers in 13 different regions of Spain were included (2003-2005). The success rate was measured by a six-month continued abstinence rate at the one-year follow-up. A logistic mixed-effects regression model, taking Basic Care Units as random-effect parameter, was performed in order to analyze gender as a predictor of smoking cessation. Results At the one-year follow-up, the six-month continuous abstinence quit rate was 9.4% in men and 8.5% in women (p = 0.400). The logistic mixed-effects regression model showed that women did not have a higher odds of being an ex-smoker than men after the analysis was adjusted for confounders (OR adjusted = 0.9, 95% CI = 0.7-1.2). Conclusions Gender does not appear to be a predictor of smoking cessation at the one-year follow-up in individuals presenting at Primary Care Centers. ClinicalTrials.gov Identifier NCT00125905. PMID:21605389

[Hyperspectral Estimation of Apple Tree Canopy LAI Based on SVM and RF Regression].

PubMed

Han, Zhao-ying; Zhu, Xi-cun; Fang, Xian-yi; Wang, Zhuo-yuan; Wang, Ling; Zhao, Geng-Xing; Jiang, Yuan-mao

2016-03-01

Leaf area index (LAI) is the dynamic index of crop population size. Hyperspectral technology can be used to estimate apple canopy LAI rapidly and nondestructively. It can be provide a reference for monitoring the tree growing and yield estimation. The Red Fuji apple trees of full bearing fruit are the researching objects. Ninety apple trees canopies spectral reflectance and LAI values were measured by the ASD Fieldspec3 spectrometer and LAI-2200 in thirty orchards in constant two years in Qixia research area of Shandong Province. The optimal vegetation indices were selected by the method of correlation analysis of the original spectral reflectance and vegetation indices. The models of predicting the LAI were built with the multivariate regression analysis method of support vector machine (SVM) and random forest (RF). The new vegetation indices, GNDVI527, ND-VI676, RVI682, FD-NVI656 and GRVI517 and the previous two main vegetation indices, NDVI670 and NDVI705, are in accordance with LAI. In the RF regression model, the calibration set decision coefficient C-R2 of 0.920 and validation set decision coefficient V-R2 of 0.889 are higher than the SVM regression model by 0.045 and 0.033 respectively. The root mean square error of calibration set C-RMSE of 0.249, the root mean square error validation set V-RMSE of 0.236 are lower than that of the SVM regression model by 0.054 and 0.058 respectively. Relative analysis of calibrating error C-RPD and relative analysis of validation set V-RPD reached 3.363 and 2.520, 0.598 and 0.262, respectively, which were higher than the SVM regression model. The measured and predicted the scatterplot trend line slope of the calibration set and validation set C-S and V-S are close to 1. The estimation result of RF regression model is better than that of the SVM. RF regression model can be used to estimate the LAI of red Fuji apple trees in full fruit period.

Extrapolating intensified forest inventory data to the surrounding landscape using landsat

Treesearch

Evan B. Brooks; John W. Coulston; Valerie A. Thomas; Randolph H. Wynne

2015-01-01

In 2011, a collection of spatially intensified plots was established on three of the Experimental Forests and Ranges (EFRs) sites with the intent of facilitating FIA program objectives for regional extrapolation. Characteristic coefficients from harmonic regression (HR) analysis of associated Landsat stacks are used as inputs into a conditional random forests model to...

Instructional Advice, Time Advice and Learning Questions in Computer Simulations

ERIC Educational Resources Information Center

Rey, Gunter Daniel

2010-01-01

Undergraduate students (N = 97) used an introductory text and a computer simulation to learn fundamental concepts about statistical analyses (e.g., analysis of variance, regression analysis and General Linear Model). Each learner was randomly assigned to one cell of a 2 (with or without instructional advice) x 2 (with or without time advice) x 2…

Using multi-trait and random regression models to identify genetic variation in tolerance of pigs to Porcine Reproductive and Respiratory Syndrome virus

USDA-ARS?s Scientific Manuscript database

Background A host can adopt two response strategies to infection: resistance (reduce pathogen load) and tolerance (minimize impact of infection on performance). Both strategies may be under genetic control and could thus be targeted for genetic improvement. Although there is evidence in support of a...

Factors Associated with Recruitment and Screening in the Treatment for Adolescents with Depression Study (TADS)

ERIC Educational Resources Information Center

May, Diane E.; Hallin, Mary J.; Kratochvil, Christopher J.; Puumala, Susan E.; Smith, Lynette S.; Reinecke, Mark A.; Silva, Susan G.; Weller, Elizabeth B.; Vitiello, Benedetto; Breland-Noble, Alfiee; March, John S.

2007-01-01

Objective: To examine factors associated with eligibility and randomization and consider the efficiency of recruitment methods. Method: Adolescents, ages 12 to 17 years, were telephone screened (N = 2,804) followed by in-person evaluation (N = 1,088) for the Treatment for Adolescents With Depression Study. Separate logistic regression models,…

Beyond logistic regression: structural equations modelling for binary variables and its application to investigating unobserved confounders.

PubMed

Kupek, Emil

2006-03-15

Structural equation modelling (SEM) has been increasingly used in medical statistics for solving a system of related regression equations. However, a great obstacle for its wider use has been its difficulty in handling categorical variables within the framework of generalised linear models. A large data set with a known structure among two related outcomes and three independent variables was generated to investigate the use of Yule's transformation of odds ratio (OR) into Q-metric by (OR-1)/(OR+1) to approximate Pearson's correlation coefficients between binary variables whose covariance structure can be further analysed by SEM. Percent of correctly classified events and non-events was compared with the classification obtained by logistic regression. The performance of SEM based on Q-metric was also checked on a small (N = 100) random sample of the data generated and on a real data set. SEM successfully recovered the generated model structure. SEM of real data suggested a significant influence of a latent confounding variable which would have not been detectable by standard logistic regression. SEM classification performance was broadly similar to that of the logistic regression. The analysis of binary data can be greatly enhanced by Yule's transformation of odds ratios into estimated correlation matrix that can be further analysed by SEM. The interpretation of results is aided by expressing them as odds ratios which are the most frequently used measure of effect in medical statistics.

Thorough statistical comparison of machine learning regression models and their ensembles for sub-pixel imperviousness and imperviousness change mapping

NASA Astrophysics Data System (ADS)

Drzewiecki, Wojciech

2017-12-01

We evaluated the performance of nine machine learning regression algorithms and their ensembles for sub-pixel estimation of impervious areas coverages from Landsat imagery. The accuracy of imperviousness mapping in individual time points was assessed based on RMSE, MAE and R2. These measures were also used for the assessment of imperviousness change intensity estimations. The applicability for detection of relevant changes in impervious areas coverages at sub-pixel level was evaluated using overall accuracy, F-measure and ROC Area Under Curve. The results proved that Cubist algorithm may be advised for Landsat-based mapping of imperviousness for single dates. Stochastic gradient boosting of regression trees (GBM) may be also considered for this purpose. However, Random Forest algorithm is endorsed for both imperviousness change detection and mapping of its intensity. In all applications the heterogeneous model ensembles performed at least as well as the best individual models or better. They may be recommended for improving the quality of sub-pixel imperviousness and imperviousness change mapping. The study revealed also limitations of the investigated methodology for detection of subtle changes of imperviousness inside the pixel. None of the tested approaches was able to reliably classify changed and non-changed pixels if the relevant change threshold was set as one or three percent. Also for fi ve percent change threshold most of algorithms did not ensure that the accuracy of change map is higher than the accuracy of random classifi er. For the threshold of relevant change set as ten percent all approaches performed satisfactory.

Accurate Segmentation of CT Male Pelvic Organs via Regression-based Deformable Models and Multi-task Random Forests

PubMed Central

Gao, Yaozong; Shao, Yeqin; Lian, Jun; Wang, Andrew Z.; Chen, Ronald C.

2016-01-01

Segmenting male pelvic organs from CT images is a prerequisite for prostate cancer radiotherapy. The efficacy of radiation treatment highly depends on segmentation accuracy. However, accurate segmentation of male pelvic organs is challenging due to low tissue contrast of CT images, as well as large variations of shape and appearance of the pelvic organs. Among existing segmentation methods, deformable models are the most popular, as shape prior can be easily incorporated to regularize the segmentation. Nonetheless, the sensitivity to initialization often limits their performance, especially for segmenting organs with large shape variations. In this paper, we propose a novel approach to guide deformable models, thus making them robust against arbitrary initializations. Specifically, we learn a displacement regressor, which predicts 3D displacement from any image voxel to the target organ boundary based on the local patch appearance. This regressor provides a nonlocal external force for each vertex of deformable model, thus overcoming the initialization problem suffered by the traditional deformable models. To learn a reliable displacement regressor, two strategies are particularly proposed. 1) A multi-task random forest is proposed to learn the displacement regressor jointly with the organ classifier; 2) an auto-context model is used to iteratively enforce structural information during voxel-wise prediction. Extensive experiments on 313 planning CT scans of 313 patients show that our method achieves better results than alternative classification or regression based methods, and also several other existing methods in CT pelvic organ segmentation. PMID:26800531

Improving virtual screening predictive accuracy of Human kallikrein 5 inhibitors using machine learning models.

PubMed

Fang, Xingang; Bagui, Sikha; Bagui, Subhash

2017-08-01

The readily available high throughput screening (HTS) data from the PubChem database provides an opportunity for mining of small molecules in a variety of biological systems using machine learning techniques. From the thousands of available molecular descriptors developed to encode useful chemical information representing the characteristics of molecules, descriptor selection is an essential step in building an optimal quantitative structural-activity relationship (QSAR) model. For the development of a systematic descriptor selection strategy, we need the understanding of the relationship between: (i) the descriptor selection; (ii) the choice of the machine learning model; and (iii) the characteristics of the target bio-molecule. In this work, we employed the Signature descriptor to generate a dataset on the Human kallikrein 5 (hK 5) inhibition confirmatory assay data and compared multiple classification models including logistic regression, support vector machine, random forest and k-nearest neighbor. Under optimal conditions, the logistic regression model provided extremely high overall accuracy (98%) and precision (90%), with good sensitivity (65%) in the cross validation test. In testing the primary HTS screening data with more than 200K molecular structures, the logistic regression model exhibited the capability of eliminating more than 99.9% of the inactive structures. As part of our exploration of the descriptor-model-target relationship, the excellent predictive performance of the combination of the Signature descriptor and the logistic regression model on the assay data of the Human kallikrein 5 (hK 5) target suggested a feasible descriptor/model selection strategy on similar targets. Copyright © 2017 Elsevier Ltd. All rights reserved.

Baseline adjustments for binary data in repeated cross-sectional cluster randomized trials.

PubMed

Nixon, R M; Thompson, S G

2003-09-15

Analysis of covariance models, which adjust for a baseline covariate, are often used to compare treatment groups in a controlled trial in which individuals are randomized. Such analysis adjusts for any baseline imbalance and usually increases the precision of the treatment effect estimate. We assess the value of such adjustments in the context of a cluster randomized trial with repeated cross-sectional design and a binary outcome. In such a design, a new sample of individuals is taken from the clusters at each measurement occasion, so that baseline adjustment has to be at the cluster level. Logistic regression models are used to analyse the data, with cluster level random effects to allow for different outcome probabilities in each cluster. We compare the estimated treatment effect and its precision in models that incorporate a covariate measuring the cluster level probabilities at baseline and those that do not. In two data sets, taken from a cluster randomized trial in the treatment of menorrhagia, the value of baseline adjustment is only evident when the number of subjects per cluster is large. We assess the generalizability of these findings by undertaking a simulation study, and find that increased precision of the treatment effect requires both large cluster sizes and substantial heterogeneity between clusters at baseline, but baseline imbalance arising by chance in a randomized study can always be effectively adjusted for. Copyright 2003 John Wiley & Sons, Ltd.

Models for Estimating Genetic Parameters of Milk Production Traits Using Random Regression Models in Korean Holstein Cattle

PubMed Central

Cho, C. I.; Alam, M.; Choi, T. J.; Choy, Y. H.; Choi, J. G.; Lee, S. S.; Cho, K. H.

2016-01-01

The objectives of the study were to estimate genetic parameters for milk production traits of Holstein cattle using random regression models (RRMs), and to compare the goodness of fit of various RRMs with homogeneous and heterogeneous residual variances. A total of 126,980 test-day milk production records of the first parity Holstein cows between 2007 and 2014 from the Dairy Cattle Improvement Center of National Agricultural Cooperative Federation in South Korea were used. These records included milk yield (MILK), fat yield (FAT), protein yield (PROT), and solids-not-fat yield (SNF). The statistical models included random effects of genetic and permanent environments using Legendre polynomials (LP) of the third to fifth order (L3–L5), fixed effects of herd-test day, year-season at calving, and a fixed regression for the test-day record (third to fifth order). The residual variances in the models were either homogeneous (HOM) or heterogeneous (15 classes, HET15; 60 classes, HET60). A total of nine models (3 orders of polynomials×3 types of residual variance) including L3-HOM, L3-HET15, L3-HET60, L4-HOM, L4-HET15, L4-HET60, L5-HOM, L5-HET15, and L5-HET60 were compared using Akaike information criteria (AIC) and/or Schwarz Bayesian information criteria (BIC) statistics to identify the model(s) of best fit for their respective traits. The lowest BIC value was observed for the models L5-HET15 (MILK; PROT; SNF) and L4-HET15 (FAT), which fit the best. In general, the BIC values of HET15 models for a particular polynomial order was lower than that of the HET60 model in most cases. This implies that the orders of LP and types of residual variances affect the goodness of models. Also, the heterogeneity of residual variances should be considered for the test-day analysis. The heritability estimates of from the best fitted models ranged from 0.08 to 0.15 for MILK, 0.06 to 0.14 for FAT, 0.08 to 0.12 for PROT, and 0.07 to 0.13 for SNF according to days in milk of first lactation. Genetic variances for studied traits tended to decrease during the earlier stages of lactation, which were followed by increases in the middle and decreases further at the end of lactation. With regards to the fitness of the models and the differential genetic parameters across the lactation stages, we could estimate genetic parameters more accurately from RRMs than from lactation models. Therefore, we suggest using RRMs in place of lactation models to make national dairy cattle genetic evaluations for milk production traits in Korea. PMID:26954184

Models for Estimating Genetic Parameters of Milk Production Traits Using Random Regression Models in Korean Holstein Cattle.

PubMed

Cho, C I; Alam, M; Choi, T J; Choy, Y H; Choi, J G; Lee, S S; Cho, K H

2016-05-01

The objectives of the study were to estimate genetic parameters for milk production traits of Holstein cattle using random regression models (RRMs), and to compare the goodness of fit of various RRMs with homogeneous and heterogeneous residual variances. A total of 126,980 test-day milk production records of the first parity Holstein cows between 2007 and 2014 from the Dairy Cattle Improvement Center of National Agricultural Cooperative Federation in South Korea were used. These records included milk yield (MILK), fat yield (FAT), protein yield (PROT), and solids-not-fat yield (SNF). The statistical models included random effects of genetic and permanent environments using Legendre polynomials (LP) of the third to fifth order (L3-L5), fixed effects of herd-test day, year-season at calving, and a fixed regression for the test-day record (third to fifth order). The residual variances in the models were either homogeneous (HOM) or heterogeneous (15 classes, HET15; 60 classes, HET60). A total of nine models (3 orders of polynomials×3 types of residual variance) including L3-HOM, L3-HET15, L3-HET60, L4-HOM, L4-HET15, L4-HET60, L5-HOM, L5-HET15, and L5-HET60 were compared using Akaike information criteria (AIC) and/or Schwarz Bayesian information criteria (BIC) statistics to identify the model(s) of best fit for their respective traits. The lowest BIC value was observed for the models L5-HET15 (MILK; PROT; SNF) and L4-HET15 (FAT), which fit the best. In general, the BIC values of HET15 models for a particular polynomial order was lower than that of the HET60 model in most cases. This implies that the orders of LP and types of residual variances affect the goodness of models. Also, the heterogeneity of residual variances should be considered for the test-day analysis. The heritability estimates of from the best fitted models ranged from 0.08 to 0.15 for MILK, 0.06 to 0.14 for FAT, 0.08 to 0.12 for PROT, and 0.07 to 0.13 for SNF according to days in milk of first lactation. Genetic variances for studied traits tended to decrease during the earlier stages of lactation, which were followed by increases in the middle and decreases further at the end of lactation. With regards to the fitness of the models and the differential genetic parameters across the lactation stages, we could estimate genetic parameters more accurately from RRMs than from lactation models. Therefore, we suggest using RRMs in place of lactation models to make national dairy cattle genetic evaluations for milk production traits in Korea.

Improving Riverine Constituent Concentration and Flux Estimation by Accounting for Antecedent Discharge Conditions

NASA Astrophysics Data System (ADS)

Zhang, Q.; Ball, W. P.

2016-12-01

Regression-based approaches are often employed to estimate riverine constituent concentrations and fluxes based on typically sparse concentration observations. One such approach is the WRTDS ("Weighted Regressions on Time, Discharge, and Season") method, which has been shown to provide more accurate estimates than prior approaches. Centered on WRTDS, this work was aimed at developing improved models for constituent concentration and flux estimation by accounting for antecedent discharge conditions. Twelve modified models were developed and tested, each of which contains one additional variable to represent antecedent conditions. High-resolution ( daily) data at nine monitoring sites were used to evaluate the relative merits of the models for estimation of six constituents - chloride (Cl), nitrate-plus-nitrite (NOx), total Kjeldahl nitrogen (TKN), total phosphorus (TP), soluble reactive phosphorus (SRP), and suspended sediment (SS). For each site-constituent combination, 30 concentration subsets were generated from the original data through Monte Carlo sub-sampling and then used to evaluate model performance. For the sub-sampling, three sampling strategies were adopted: (A) 1 random sample each month (12/year), (B) 12 random monthly samples plus additional 8 random samples per year (20/year), and (C) 12 regular (non-storm) and 8 storm samples per year (20/year). The modified models show general improvement over the original model under all three sampling strategies. Major improvements were achieved for NOx by the long-term flow-anomaly model and for Cl by the ADF (average discounted flow) model and the short-term flow-anomaly model. Moderate improvements were achieved for SS, TP, and TKN by the ADF model. By contrast, no such achievement was achieved for SRP by any proposed model. In terms of sampling strategy, performance of all models was generally best using strategy C and worst using strategy A, and especially so for SS, TP, and SRP, confirming the value of routinely collecting storm-flow samples. Overall, this work provides a comprehensive set of statistical evidence for supporting the incorporation of antecedent discharge conditions into WRTDS for constituent concentration and flux estimation, thereby combining the advantages of two recent developments in water quality modeling.

Genetic parameters for test day milk yields of first lactation Holstein cows by random regression models.

PubMed

de Melo, C M R; Packer, I U; Costa, C N; Machado, P F

2007-03-01

Covariance components for test day milk yield using 263 390 first lactation records of 32 448 Holstein cows were estimated using random regression animal models by restricted maximum likelihood. Three functions were used to adjust the lactation curve: the five-parameter logarithmic Ali and Schaeffer function (AS), the three-parameter exponential Wilmink function in its standard form (W) and in a modified form (W*), by reducing the range of covariate, and the combination of Legendre polynomial and W (LEG+W). Heterogeneous residual variance (RV) for different classes (4 and 29) of days in milk was considered in adjusting the functions. Estimates of RV were quite similar, rating from 4.15 to 5.29 kg2. Heritability estimates for AS (0.29 to 0.42), LEG+W (0.28 to 0.42) and W* (0.33 to 0.40) were similar, but heritability estimates used W (0.25 to 0.65) were highest than those estimated by the other functions, particularly at the end of lactation. Genetic correlations between milk yield on consecutive test days were close to unity, but decreased as the interval between test days increased. The AS function with homogeneous RV model had the best fit among those evaluated.

A closer look at cross-validation for assessing the accuracy of gene regulatory networks and models.

PubMed

Tabe-Bordbar, Shayan; Emad, Amin; Zhao, Sihai Dave; Sinha, Saurabh

2018-04-26

Cross-validation (CV) is a technique to assess the generalizability of a model to unseen data. This technique relies on assumptions that may not be satisfied when studying genomics datasets. For example, random CV (RCV) assumes that a randomly selected set of samples, the test set, well represents unseen data. This assumption doesn't hold true where samples are obtained from different experimental conditions, and the goal is to learn regulatory relationships among the genes that generalize beyond the observed conditions. In this study, we investigated how the CV procedure affects the assessment of supervised learning methods used to learn gene regulatory networks (or in other applications). We compared the performance of a regression-based method for gene expression prediction estimated using RCV with that estimated using a clustering-based CV (CCV) procedure. Our analysis illustrates that RCV can produce over-optimistic estimates of the model's generalizability compared to CCV. Next, we defined the 'distinctness' of test set from training set and showed that this measure is predictive of performance of the regression method. Finally, we introduced a simulated annealing method to construct partitions with gradually increasing distinctness and showed that performance of different gene expression prediction methods can be better evaluated using this method.

Neural network modeling for surgical decisions on traumatic brain injury patients.

PubMed

Li, Y C; Liu, L; Chiu, W T; Jian, W S

2000-01-01

Computerized medical decision support systems have been a major research topic in recent years. Intelligent computer programs were implemented to aid physicians and other medical professionals in making difficult medical decisions. This report compares three different mathematical models for building a traumatic brain injury (TBI) medical decision support system (MDSS). These models were developed based on a large TBI patient database. This MDSS accepts a set of patient data such as the types of skull fracture, Glasgow Coma Scale (GCS), episode of convulsion and return the chance that a neurosurgeon would recommend an open-skull surgery for this patient. The three mathematical models described in this report including a logistic regression model, a multi-layer perceptron (MLP) neural network and a radial-basis-function (RBF) neural network. From the 12,640 patients selected from the database. A randomly drawn 9480 cases were used as the training group to develop/train our models. The other 3160 cases were in the validation group which we used to evaluate the performance of these models. We used sensitivity, specificity, areas under receiver-operating characteristics (ROC) curve and calibration curves as the indicator of how accurate these models are in predicting a neurosurgeon's decision on open-skull surgery. The results showed that, assuming equal importance of sensitivity and specificity, the logistic regression model had a (sensitivity, specificity) of (73%, 68%), compared to (80%, 80%) from the RBF model and (88%, 80%) from the MLP model. The resultant areas under ROC curve for logistic regression, RBF and MLP neural networks are 0.761, 0.880 and 0.897, respectively (P < 0.05). Among these models, the logistic regression has noticeably poorer calibration. This study demonstrated the feasibility of applying neural networks as the mechanism for TBI decision support systems based on clinical databases. The results also suggest that neural networks may be a better solution for complex, non-linear medical decision support systems than conventional statistical techniques such as logistic regression.

Random forests as cumulative effects models: A case study of lakes and rivers in Muskoka, Canada.

PubMed

Jones, F Chris; Plewes, Rachel; Murison, Lorna; MacDougall, Mark J; Sinclair, Sarah; Davies, Christie; Bailey, John L; Richardson, Murray; Gunn, John

2017-10-01

Cumulative effects assessment (CEA) - a type of environmental appraisal - lacks effective methods for modeling cumulative effects, evaluating indicators of ecosystem condition, and exploring the likely outcomes of development scenarios. Random forests are an extension of classification and regression trees, which model response variables by recursive partitioning. Random forests were used to model a series of candidate ecological indicators that described lakes and rivers from a case study watershed (The Muskoka River Watershed, Canada). Suitability of the candidate indicators for use in cumulative effects assessment and watershed monitoring was assessed according to how well they could be predicted from natural habitat features and how sensitive they were to human land-use. The best models explained 75% of the variation in a multivariate descriptor of lake benthic-macroinvertebrate community structure, and 76% of the variation in the conductivity of river water. Similar results were obtained by cross-validation. Several candidate indicators detected a simulated doubling of urban land-use in their catchments, and a few were able to detect a simulated doubling of agricultural land-use. The paper demonstrates that random forests can be used to describe the combined and singular effects of multiple stressors and natural environmental factors, and furthermore, that random forests can be used to evaluate the performance of monitoring indicators. The numerical methods presented are applicable to any ecosystem and indicator type, and therefore represent a step forward for CEA. Crown Copyright © 2017. Published by Elsevier Ltd. All rights reserved.

Exploring prediction uncertainty of spatial data in geostatistical and machine learning Approaches

NASA Astrophysics Data System (ADS)

Klump, J. F.; Fouedjio, F.

2017-12-01

Geostatistical methods such as kriging with external drift as well as machine learning techniques such as quantile regression forest have been intensively used for modelling spatial data. In addition to providing predictions for target variables, both approaches are able to deliver a quantification of the uncertainty associated with the prediction at a target location. Geostatistical approaches are, by essence, adequate for providing such prediction uncertainties and their behaviour is well understood. However, they often require significant data pre-processing and rely on assumptions that are rarely met in practice. Machine learning algorithms such as random forest regression, on the other hand, require less data pre-processing and are non-parametric. This makes the application of machine learning algorithms to geostatistical problems an attractive proposition. The objective of this study is to compare kriging with external drift and quantile regression forest with respect to their ability to deliver reliable prediction uncertainties of spatial data. In our comparison we use both simulated and real world datasets. Apart from classical performance indicators, comparisons make use of accuracy plots, probability interval width plots, and the visual examinations of the uncertainty maps provided by the two approaches. By comparing random forest regression to kriging we found that both methods produced comparable maps of estimated values for our variables of interest. However, the measure of uncertainty provided by random forest seems to be quite different to the measure of uncertainty provided by kriging. In particular, the lack of spatial context can give misleading results in areas without ground truth data. These preliminary results raise questions about assessing the risks associated with decisions based on the predictions from geostatistical and machine learning algorithms in a spatial context, e.g. mineral exploration.

Genetic Parameters for Milk Yield and Lactation Persistency Using Random Regression Models in Girolando Cattle

PubMed Central

Canaza-Cayo, Ali William; Lopes, Paulo Sávio; da Silva, Marcos Vinicius Gualberto Barbosa; de Almeida Torres, Robledo; Martins, Marta Fonseca; Arbex, Wagner Antonio; Cobuci, Jaime Araujo

2015-01-01

A total of 32,817 test-day milk yield (TDMY) records of the first lactation of 4,056 Girolando cows daughters of 276 sires, collected from 118 herds between 2000 and 2011 were utilized to estimate the genetic parameters for TDMY via random regression models (RRM) using Legendre’s polynomial functions whose orders varied from 3 to 5. In addition, nine measures of persistency in milk yield (PSi) and the genetic trend of 305-day milk yield (305MY) were evaluated. The fit quality criteria used indicated RRM employing the Legendre’s polynomial of orders 3 and 5 for fitting the genetic additive and permanent environment effects, respectively, as the best model. The heritability and genetic correlation for TDMY throughout the lactation, obtained with the best model, varied from 0.18 to 0.23 and from −0.03 to 1.00, respectively. The heritability and genetic correlation for persistency and 305MY varied from 0.10 to 0.33 and from −0.98 to 1.00, respectively. The use of PS7 would be the most suitable option for the evaluation of Girolando cattle. The estimated breeding values for 305MY of sires and cows showed significant and positive genetic trends. Thus, the use of selection indices would be indicated in the genetic evaluation of Girolando cattle for both traits. PMID:26323397

Variable Selection for Regression Models of Percentile Flows

NASA Astrophysics Data System (ADS)

Fouad, G.

2017-12-01

Percentile flows describe the flow magnitude equaled or exceeded for a given percent of time, and are widely used in water resource management. However, these statistics are normally unavailable since most basins are ungauged. Percentile flows of ungauged basins are often predicted using regression models based on readily observable basin characteristics, such as mean elevation. The number of these independent variables is too large to evaluate all possible models. A subset of models is typically evaluated using automatic procedures, like stepwise regression. This ignores a large variety of methods from the field of feature (variable) selection and physical understanding of percentile flows. A study of 918 basins in the United States was conducted to compare an automatic regression procedure to the following variable selection methods: (1) principal component analysis, (2) correlation analysis, (3) random forests, (4) genetic programming, (5) Bayesian networks, and (6) physical understanding. The automatic regression procedure only performed better than principal component analysis. Poor performance of the regression procedure was due to a commonly used filter for multicollinearity, which rejected the strongest models because they had cross-correlated independent variables. Multicollinearity did not decrease model performance in validation because of a representative set of calibration basins. Variable selection methods based strictly on predictive power (numbers 2-5 from above) performed similarly, likely indicating a limit to the predictive power of the variables. Similar performance was also reached using variables selected based on physical understanding, a finding that substantiates recent calls to emphasize physical understanding in modeling for predictions in ungauged basins. The strongest variables highlighted the importance of geology and land cover, whereas widely used topographic variables were the weakest predictors. Variables suffered from a high degree of multicollinearity, possibly illustrating the co-evolution of climatic and physiographic conditions. Given the ineffectiveness of many variables used here, future work should develop new variables that target specific processes associated with percentile flows.

Application of random effects to the study of resource selection by animals

USGS Publications Warehouse

Gillies, C.S.; Hebblewhite, M.; Nielsen, S.E.; Krawchuk, M.A.; Aldridge, Cameron L.; Frair, J.L.; Saher, D.J.; Stevens, C.E.; Jerde, C.L.

2006-01-01

1. Resource selection estimated by logistic regression is used increasingly in studies to identify critical resources for animal populations and to predict species occurrence.2. Most frequently, individual animals are monitored and pooled to estimate population-level effects without regard to group or individual-level variation. Pooling assumes that both observations and their errors are independent, and resource selection is constant given individual variation in resource availability.3. Although researchers have identified ways to minimize autocorrelation, variation between individuals caused by differences in selection or available resources, including functional responses in resource selection, have not been well addressed.4. Here we review random-effects models and their application to resource selection modelling to overcome these common limitations. We present a simple case study of an analysis of resource selection by grizzly bears in the foothills of the Canadian Rocky Mountains with and without random effects.5. Both categorical and continuous variables in the grizzly bear model differed in interpretation, both in statistical significance and coefficient sign, depending on how a random effect was included. We used a simulation approach to clarify the application of random effects under three common situations for telemetry studies: (a) discrepancies in sample sizes among individuals; (b) differences among individuals in selection where availability is constant; and (c) differences in availability with and without a functional response in resource selection.6. We found that random intercepts accounted for unbalanced sample designs, and models with random intercepts and coefficients improved model fit given the variation in selection among individuals and functional responses in selection. Our empirical example and simulations demonstrate how including random effects in resource selection models can aid interpretation and address difficult assumptions limiting their generality. This approach will allow researchers to appropriately estimate marginal (population) and conditional (individual) responses, and account for complex grouping, unbalanced sample designs and autocorrelation.

Application of random effects to the study of resource selection by animals.

PubMed

Gillies, Cameron S; Hebblewhite, Mark; Nielsen, Scott E; Krawchuk, Meg A; Aldridge, Cameron L; Frair, Jacqueline L; Saher, D Joanne; Stevens, Cameron E; Jerde, Christopher L

2006-07-01

1. Resource selection estimated by logistic regression is used increasingly in studies to identify critical resources for animal populations and to predict species occurrence. 2. Most frequently, individual animals are monitored and pooled to estimate population-level effects without regard to group or individual-level variation. Pooling assumes that both observations and their errors are independent, and resource selection is constant given individual variation in resource availability. 3. Although researchers have identified ways to minimize autocorrelation, variation between individuals caused by differences in selection or available resources, including functional responses in resource selection, have not been well addressed. 4. Here we review random-effects models and their application to resource selection modelling to overcome these common limitations. We present a simple case study of an analysis of resource selection by grizzly bears in the foothills of the Canadian Rocky Mountains with and without random effects. 5. Both categorical and continuous variables in the grizzly bear model differed in interpretation, both in statistical significance and coefficient sign, depending on how a random effect was included. We used a simulation approach to clarify the application of random effects under three common situations for telemetry studies: (a) discrepancies in sample sizes among individuals; (b) differences among individuals in selection where availability is constant; and (c) differences in availability with and without a functional response in resource selection. 6. We found that random intercepts accounted for unbalanced sample designs, and models with random intercepts and coefficients improved model fit given the variation in selection among individuals and functional responses in selection. Our empirical example and simulations demonstrate how including random effects in resource selection models can aid interpretation and address difficult assumptions limiting their generality. This approach will allow researchers to appropriately estimate marginal (population) and conditional (individual) responses, and account for complex grouping, unbalanced sample designs and autocorrelation.

Assessing the predictive capability of randomized tree-based ensembles in streamflow modelling

NASA Astrophysics Data System (ADS)

Galelli, S.; Castelletti, A.

2013-02-01

Combining randomization methods with ensemble prediction is emerging as an effective option to balance accuracy and computational efficiency in data-driven modeling. In this paper we investigate the prediction capability of extremely randomized trees (Extra-Trees), in terms of accuracy, explanation ability and computational efficiency, in a streamflow modeling exercise. Extra-Trees are a totally randomized tree-based ensemble method that (i) alleviates the poor generalization property and tendency to overfitting of traditional standalone decision trees (e.g. CART); (ii) is computationally very efficient; and, (iii) allows to infer the relative importance of the input variables, which might help in the ex-post physical interpretation of the model. The Extra-Trees potential is analyzed on two real-world case studies (Marina catchment (Singapore) and Canning River (Western Australia)) representing two different morphoclimatic contexts comparatively with other tree-based methods (CART and M5) and parametric data-driven approaches (ANNs and multiple linear regression). Results show that Extra-Trees perform comparatively well to the best of the benchmarks (i.e. M5) in both the watersheds, while outperforming the other approaches in terms of computational requirement when adopted on large datasets. In addition, the ranking of the input variable provided can be given a physically meaningful interpretation.

Assessing the predictive capability of randomized tree-based ensembles in streamflow modelling

NASA Astrophysics Data System (ADS)

Galelli, S.; Castelletti, A.

2013-07-01

Combining randomization methods with ensemble prediction is emerging as an effective option to balance accuracy and computational efficiency in data-driven modelling. In this paper, we investigate the prediction capability of extremely randomized trees (Extra-Trees), in terms of accuracy, explanation ability and computational efficiency, in a streamflow modelling exercise. Extra-Trees are a totally randomized tree-based ensemble method that (i) alleviates the poor generalisation property and tendency to overfitting of traditional standalone decision trees (e.g. CART); (ii) is computationally efficient; and, (iii) allows to infer the relative importance of the input variables, which might help in the ex-post physical interpretation of the model. The Extra-Trees potential is analysed on two real-world case studies - Marina catchment (Singapore) and Canning River (Western Australia) - representing two different morphoclimatic contexts. The evaluation is performed against other tree-based methods (CART and M5) and parametric data-driven approaches (ANNs and multiple linear regression). Results show that Extra-Trees perform comparatively well to the best of the benchmarks (i.e. M5) in both the watersheds, while outperforming the other approaches in terms of computational requirement when adopted on large datasets. In addition, the ranking of the input variable provided can be given a physically meaningful interpretation.

Risk factors for highly pathogenic avian influenza in commercial layer chicken farms in bangladesh during 2011.

PubMed

Osmani, M G; Thornton, R N; Dhand, N K; Hoque, M A; Milon, Sk M A; Kalam, M A; Hossain, M; Yamage, M

2014-12-01

A case-control study conducted during 2011 involved 90 randomly selected commercial layer farms infected with highly pathogenic avian influenza type A subtype H5N1 (HPAI) and 175 control farms randomly selected from within 5 km of infected farms. A questionnaire was designed to obtain information about potential risk factors for contracting HPAI and was administered to farm owners or managers. Logistic regression analyses were conducted to identify significant risk factors. A total of 20 of 43 risk factors for contracting HPAI were identified after univariable logistic regression analysis. A multivariable logistic regression model was derived by forward stepwise selection. Both unmatched and matched analyses were performed. The key risk factors identified were numbers of staff, frequency of veterinary visits, presence of village chickens roaming on the farm and staff trading birds. Aggregating these findings with those from other studies resulted in a list of 16 key risk factors identified in Bangladesh. Most of these related to biosecurity. It is considered feasible for Bangladesh to achieve a very low incidence of HPAI. Using the cumulative list of risk factors to enhance biosecurity pertaining to commercial farms would facilitate this objective. © 2013 Blackwell Verlag GmbH.

The median hazard ratio: a useful measure of variance and general contextual effects in multilevel survival analysis.

PubMed

Austin, Peter C; Wagner, Philippe; Merlo, Juan

2017-03-15

Multilevel data occurs frequently in many research areas like health services research and epidemiology. A suitable way to analyze such data is through the use of multilevel regression models (MLRM). MLRM incorporate cluster-specific random effects which allow one to partition the total individual variance into between-cluster variation and between-individual variation. Statistically, MLRM account for the dependency of the data within clusters and provide correct estimates of uncertainty around regression coefficients. Substantively, the magnitude of the effect of clustering provides a measure of the General Contextual Effect (GCE). When outcomes are binary, the GCE can also be quantified by measures of heterogeneity like the Median Odds Ratio (MOR) calculated from a multilevel logistic regression model. Time-to-event outcomes within a multilevel structure occur commonly in epidemiological and medical research. However, the Median Hazard Ratio (MHR) that corresponds to the MOR in multilevel (i.e., 'frailty') Cox proportional hazards regression is rarely used. Analogously to the MOR, the MHR is the median relative change in the hazard of the occurrence of the outcome when comparing identical subjects from two randomly selected different clusters that are ordered by risk. We illustrate the application and interpretation of the MHR in a case study analyzing the hazard of mortality in patients hospitalized for acute myocardial infarction at hospitals in Ontario, Canada. We provide R code for computing the MHR. The MHR is a useful and intuitive measure for expressing cluster heterogeneity in the outcome and, thereby, estimating general contextual effects in multilevel survival analysis. © 2016 The Authors. Statistics in Medicine published by John Wiley & Sons Ltd. © 2016 The Authors. Statistics in Medicine published by John Wiley & Sons Ltd.

The median hazard ratio: a useful measure of variance and general contextual effects in multilevel survival analysis

PubMed Central

Wagner, Philippe; Merlo, Juan

2016-01-01

Multilevel data occurs frequently in many research areas like health services research and epidemiology. A suitable way to analyze such data is through the use of multilevel regression models (MLRM). MLRM incorporate cluster‐specific random effects which allow one to partition the total individual variance into between‐cluster variation and between‐individual variation. Statistically, MLRM account for the dependency of the data within clusters and provide correct estimates of uncertainty around regression coefficients. Substantively, the magnitude of the effect of clustering provides a measure of the General Contextual Effect (GCE). When outcomes are binary, the GCE can also be quantified by measures of heterogeneity like the Median Odds Ratio (MOR) calculated from a multilevel logistic regression model. Time‐to‐event outcomes within a multilevel structure occur commonly in epidemiological and medical research. However, the Median Hazard Ratio (MHR) that corresponds to the MOR in multilevel (i.e., ‘frailty’) Cox proportional hazards regression is rarely used. Analogously to the MOR, the MHR is the median relative change in the hazard of the occurrence of the outcome when comparing identical subjects from two randomly selected different clusters that are ordered by risk. We illustrate the application and interpretation of the MHR in a case study analyzing the hazard of mortality in patients hospitalized for acute myocardial infarction at hospitals in Ontario, Canada. We provide R code for computing the MHR. The MHR is a useful and intuitive measure for expressing cluster heterogeneity in the outcome and, thereby, estimating general contextual effects in multilevel survival analysis. © 2016 The Authors. Statistics in Medicine published by John Wiley & Sons Ltd. PMID:27885709

The Naïve Overfitting Index Selection (NOIS): A new method to optimize model complexity for hyperspectral data

NASA Astrophysics Data System (ADS)

Rocha, Alby D.; Groen, Thomas A.; Skidmore, Andrew K.; Darvishzadeh, Roshanak; Willemen, Louise

2017-11-01

The growing number of narrow spectral bands in hyperspectral remote sensing improves the capacity to describe and predict biological processes in ecosystems. But it also poses a challenge to fit empirical models based on such high dimensional data, which often contain correlated and noisy predictors. As sample sizes, to train and validate empirical models, seem not to be increasing at the same rate, overfitting has become a serious concern. Overly complex models lead to overfitting by capturing more than the underlying relationship, and also through fitting random noise in the data. Many regression techniques claim to overcome these problems by using different strategies to constrain complexity, such as limiting the number of terms in the model, by creating latent variables or by shrinking parameter coefficients. This paper is proposing a new method, named Naïve Overfitting Index Selection (NOIS), which makes use of artificially generated spectra, to quantify the relative model overfitting and to select an optimal model complexity supported by the data. The robustness of this new method is assessed by comparing it to a traditional model selection based on cross-validation. The optimal model complexity is determined for seven different regression techniques, such as partial least squares regression, support vector machine, artificial neural network and tree-based regressions using five hyperspectral datasets. The NOIS method selects less complex models, which present accuracies similar to the cross-validation method. The NOIS method reduces the chance of overfitting, thereby avoiding models that present accurate predictions that are only valid for the data used, and too complex to make inferences about the underlying process.

Local linear regression for function learning: an analysis based on sample discrepancy.

PubMed

Cervellera, Cristiano; Macciò, Danilo

2014-11-01

Local linear regression models, a kind of nonparametric structures that locally perform a linear estimation of the target function, are analyzed in the context of empirical risk minimization (ERM) for function learning. The analysis is carried out with emphasis on geometric properties of the available data. In particular, the discrepancy of the observation points used both to build the local regression models and compute the empirical risk is considered. This allows to treat indifferently the case in which the samples come from a random external source and the one in which the input space can be freely explored. Both consistency of the ERM procedure and approximating capabilities of the estimator are analyzed, proving conditions to ensure convergence. Since the theoretical analysis shows that the estimation improves as the discrepancy of the observation points becomes smaller, low-discrepancy sequences, a family of sampling methods commonly employed for efficient numerical integration, are also analyzed. Simulation results involving two different examples of function learning are provided.

Empirical Performance of Cross-Validation With Oracle Methods in a Genomics Context

PubMed Central

Martinez, Josue G.; Carroll, Raymond J.; Müller, Samuel; Sampson, Joshua N.; Chatterjee, Nilanjan

2012-01-01

When employing model selection methods with oracle properties such as the smoothly clipped absolute deviation (SCAD) and the Adaptive Lasso, it is typical to estimate the smoothing parameter by m-fold cross-validation, for example, m = 10. In problems where the true regression function is sparse and the signals large, such cross-validation typically works well. However, in regression modeling of genomic studies involving Single Nucleotide Polymorphisms (SNP), the true regression functions, while thought to be sparse, do not have large signals. We demonstrate empirically that in such problems, the number of selected variables using SCAD and the Adaptive Lasso, with 10-fold cross-validation, is a random variable that has considerable and surprising variation. Similar remarks apply to non-oracle methods such as the Lasso. Our study strongly questions the suitability of performing only a single run of m-fold cross-validation with any oracle method, and not just the SCAD and Adaptive Lasso. PMID:22347720

Genetic prediction of type 2 diabetes using deep neural network.

PubMed

Kim, J; Kim, J; Kwak, M J; Bajaj, M

2018-04-01

Type 2 diabetes (T2DM) has strong heritability but genetic models to explain heritability have been challenging. We tested deep neural network (DNN) to predict T2DM using the nested case-control study of Nurses' Health Study (3326 females, 45.6% T2DM) and Health Professionals Follow-up Study (2502 males, 46.5% T2DM). We selected 96, 214, 399, and 678 single-nucleotide polymorphism (SNPs) through Fisher's exact test and L1-penalized logistic regression. We split each dataset randomly in 4:1 to train prediction models and test their performance. DNN and logistic regressions showed better area under the curve (AUC) of ROC curves than the clinical model when 399 or more SNPs included. DNN was superior than logistic regressions in AUC with 399 or more SNPs in male and 678 SNPs in female. Addition of clinical factors consistently increased AUC of DNN but failed to improve logistic regressions with 214 or more SNPs. In conclusion, we show that DNN can be a versatile tool to predict T2DM incorporating large numbers of SNPs and clinical information. Limitations include a relatively small number of the subjects mostly of European ethnicity. Further studies are warranted to confirm and improve performance of genetic prediction models using DNN in different ethnic groups. © 2017 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.

Conformal Regression for Quantitative Structure-Activity Relationship Modeling-Quantifying Prediction Uncertainty.

PubMed

Svensson, Fredrik; Aniceto, Natalia; Norinder, Ulf; Cortes-Ciriano, Isidro; Spjuth, Ola; Carlsson, Lars; Bender, Andreas

2018-05-29

Making predictions with an associated confidence is highly desirable as it facilitates decision making and resource prioritization. Conformal regression is a machine learning framework that allows the user to define the required confidence and delivers predictions that are guaranteed to be correct to the selected extent. In this study, we apply conformal regression to model molecular properties and bioactivity values and investigate different ways to scale the resultant prediction intervals to create as efficient (i.e., narrow) regressors as possible. Different algorithms to estimate the prediction uncertainty were used to normalize the prediction ranges, and the different approaches were evaluated on 29 publicly available data sets. Our results show that the most efficient conformal regressors are obtained when using the natural exponential of the ensemble standard deviation from the underlying random forest to scale the prediction intervals, but other approaches were almost as efficient. This approach afforded an average prediction range of 1.65 pIC50 units at the 80% confidence level when applied to bioactivity modeling. The choice of nonconformity function has a pronounced impact on the average prediction range with a difference of close to one log unit in bioactivity between the tightest and widest prediction range. Overall, conformal regression is a robust approach to generate bioactivity predictions with associated confidence.

Commensurate Priors for Incorporating Historical Information in Clinical Trials Using General and Generalized Linear Models

PubMed Central

Hobbs, Brian P.; Sargent, Daniel J.; Carlin, Bradley P.

2014-01-01

Assessing between-study variability in the context of conventional random-effects meta-analysis is notoriously difficult when incorporating data from only a small number of historical studies. In order to borrow strength, historical and current data are often assumed to be fully homogeneous, but this can have drastic consequences for power and Type I error if the historical information is biased. In this paper, we propose empirical and fully Bayesian modifications of the commensurate prior model (Hobbs et al., 2011) extending Pocock (1976), and evaluate their frequentist and Bayesian properties for incorporating patient-level historical data using general and generalized linear mixed regression models. Our proposed commensurate prior models lead to preposterior admissible estimators that facilitate alternative bias-variance trade-offs than those offered by pre-existing methodologies for incorporating historical data from a small number of historical studies. We also provide a sample analysis of a colon cancer trial comparing time-to-disease progression using a Weibull regression model. PMID:24795786

[Spatial differentiation and impact factors of Yutian Oasis's soil surface salt based on GWR model].

PubMed

Yuan, Yu Yun; Wahap, Halik; Guan, Jing Yun; Lu, Long Hui; Zhang, Qin Qin

2016-10-01

In this paper, topsoil salinity data gathered from 24 sampling sites in the Yutian Oasis were used, nine different kinds of environmental variables closely related to soil salinity were selec-ted as influencing factors, then, the spatial distribution characteristics of topsoil salinity and spatial heterogeneity of influencing factors were analyzed by combining the spatial autocorrelation with traditional regression analysis and geographically weighted regression model. Results showed that the topsoil salinity in Yutian Oasis was not of random distribution but had strong spatial dependence, and the spatial autocorrelation index for topsoil salinity was 0.479. Groundwater salinity, groundwater depth, elevation and temperature were the main factors influencing topsoil salt accumulation in arid land oases and they were spatially heterogeneous. The nine selected environmental variables except soil pH had significant influences on topsoil salinity with spatial disparity. GWR model was superior to the OLS model on interpretation and estimation of spatial non-stationary data, also had a remarkable advantage in visualization of modeling parameters.

Modeling Information Content Via Dirichlet-Multinomial Regression Analysis.

PubMed

Ferrari, Alberto

2017-01-01

Shannon entropy is being increasingly used in biomedical research as an index of complexity and information content in sequences of symbols, e.g. languages, amino acid sequences, DNA methylation patterns and animal vocalizations. Yet, distributional properties of information entropy as a random variable have seldom been the object of study, leading to researchers mainly using linear models or simulation-based analytical approach to assess differences in information content, when entropy is measured repeatedly in different experimental conditions. Here a method to perform inference on entropy in such conditions is proposed. Building on results coming from studies in the field of Bayesian entropy estimation, a symmetric Dirichlet-multinomial regression model, able to deal efficiently with the issue of mean entropy estimation, is formulated. Through a simulation study the model is shown to outperform linear modeling in a vast range of scenarios and to have promising statistical properties. As a practical example, the method is applied to a data set coming from a real experiment on animal communication.

A machine learning method to estimate PM2.5 concentrations across China with remote sensing, meteorological and land use information.

PubMed

Chen, Gongbo; Li, Shanshan; Knibbs, Luke D; Hamm, N A S; Cao, Wei; Li, Tiantian; Guo, Jianping; Ren, Hongyan; Abramson, Michael J; Guo, Yuming

2018-09-15

Machine learning algorithms have very high predictive ability. However, no study has used machine learning to estimate historical concentrations of PM 2.5 (particulate matter with aerodynamic diameter ≤ 2.5 μm) at daily time scale in China at a national level. To estimate daily concentrations of PM 2.5 across China during 2005-2016. Daily ground-level PM 2.5 data were obtained from 1479 stations across China during 2014-2016. Data on aerosol optical depth (AOD), meteorological conditions and other predictors were downloaded. A random forests model (non-parametric machine learning algorithms) and two traditional regression models were developed to estimate ground-level PM 2.5 concentrations. The best-fit model was then utilized to estimate the daily concentrations of PM 2.5 across China with a resolution of 0.1° (≈10 km) during 2005-2016. The daily random forests model showed much higher predictive accuracy than the other two traditional regression models, explaining the majority of spatial variability in daily PM 2.5 [10-fold cross-validation (CV) R 2  = 83%, root mean squared prediction error (RMSE) = 28.1 μg/m 3 ]. At the monthly and annual time-scale, the explained variability of average PM 2.5 increased up to 86% (RMSE = 10.7 μg/m 3 and 6.9 μg/m 3 , respectively). Taking advantage of a novel application of modeling framework and the most recent ground-level PM 2.5 observations, the machine learning method showed higher predictive ability than previous studies. Random forests approach can be used to estimate historical exposure to PM 2.5 in China with high accuracy. Copyright © 2018 Elsevier B.V. All rights reserved.

Comparison of stream invertebrate response models for bioassessment metric

USGS Publications Warehouse

Waite, Ian R.; Kennen, Jonathan G.; May, Jason T.; Brown, Larry R.; Cuffney, Thomas F.; Jones, Kimberly A.; Orlando, James L.

2012-01-01

We aggregated invertebrate data from various sources to assemble data for modeling in two ecoregions in Oregon and one in California. Our goal was to compare the performance of models developed using multiple linear regression (MLR) techniques with models developed using three relatively new techniques: classification and regression trees (CART), random forest (RF), and boosted regression trees (BRT). We used tolerance of taxa based on richness (RICHTOL) and ratio of observed to expected taxa (O/E) as response variables and land use/land cover as explanatory variables. Responses were generally linear; therefore, there was little improvement to the MLR models when compared to models using CART and RF. In general, the four modeling techniques (MLR, CART, RF, and BRT) consistently selected the same primary explanatory variables for each region. However, results from the BRT models showed significant improvement over the MLR models for each region; increases in R2 from 0.09 to 0.20. The O/E metric that was derived from models specifically calibrated for Oregon consistently had lower R2 values than RICHTOL for the two regions tested. Modeled O/E R2 values were between 0.06 and 0.10 lower for each of the four modeling methods applied in the Willamette Valley and were between 0.19 and 0.36 points lower for the Blue Mountains. As a result, BRT models may indeed represent a good alternative to MLR for modeling species distribution relative to environmental variables.

Empirical analyses of plant-climate relationships for the western United States

Treesearch

Gerald E. Rehfeldt; Nicholas L. Crookston; Marcus V. Warwell; Jeffrey S. Evans

2006-01-01

The Random Forests multiple-regression tree was used to model climate profiles of 25 biotic communities of the western United States and nine of their constituent species. Analyses of the communities were based on a gridded sample of ca. 140,000 points, while those for the species used presence-absence data from ca. 120,000 locations. Independent variables included 35...

The Influence of Neighborhood Characteristics and Parenting Practices on Academic Problems and Aggression Outcomes among Moderately to Highly Aggressive Children

ERIC Educational Resources Information Center

Barry, Tammy D.; Lochman, John E.; Fite, Paula J.; Wells, Karen C.; Colder, Craig R.

2012-01-01

The current study utilized a longitudinal design to examine the effects of neighborhood and parenting on 120 at-risk children's academic and aggressive outcomes, concurrently and at two later timepoints during the transition to middle school. Random effects regression models were estimated to examine whether neighborhood characteristics and harsh…

Machine learning to predict the occurrence of bisphosphonate-related osteonecrosis of the jaw associated with dental extraction: A preliminary report.

PubMed

Kim, Dong Wook; Kim, Hwiyoung; Nam, Woong; Kim, Hyung Jun; Cha, In-Ho

2018-04-23

The aim of this study was to build and validate five types of machine learning models that can predict the occurrence of BRONJ associated with dental extraction in patients taking bisphosphonates for the management of osteoporosis. A retrospective review of the medical records was conducted to obtain cases and controls for the study. Total 125 patients consisting of 41 cases and 84 controls were selected for the study. Five machine learning prediction algorithms including multivariable logistic regression model, decision tree, support vector machine, artificial neural network, and random forest were implemented. The outputs of these models were compared with each other and also with conventional methods, such as serum CTX level. Area under the receiver operating characteristic (ROC) curve (AUC) was used to compare the results. The performance of machine learning models was significantly superior to conventional statistical methods and single predictors. The random forest model yielded the best performance (AUC = 0.973), followed by artificial neural network (AUC = 0.915), support vector machine (AUC = 0.882), logistic regression (AUC = 0.844), decision tree (AUC = 0.821), drug holiday alone (AUC = 0.810), and CTX level alone (AUC = 0.630). Machine learning methods showed superior performance in predicting BRONJ associated with dental extraction compared to conventional statistical methods using drug holiday and serum CTX level. Machine learning can thus be applied in a wide range of clinical studies. Copyright © 2017. Published by Elsevier Inc.

Mixed models, linear dependency, and identification in age-period-cohort models.

PubMed

O'Brien, Robert M

2017-07-20

This paper examines the identification problem in age-period-cohort models that use either linear or categorically coded ages, periods, and cohorts or combinations of these parameterizations. These models are not identified using the traditional fixed effect regression model approach because of a linear dependency between the ages, periods, and cohorts. However, these models can be identified if the researcher introduces a single just identifying constraint on the model coefficients. The problem with such constraints is that the results can differ substantially depending on the constraint chosen. Somewhat surprisingly, age-period-cohort models that specify one or more of ages and/or periods and/or cohorts as random effects are identified. This is the case without introducing an additional constraint. I label this identification as statistical model identification and show how statistical model identification comes about in mixed models and why which effects are treated as fixed and which are treated as random can substantially change the estimates of the age, period, and cohort effects. Copyright © 2017 John Wiley & Sons, Ltd. Copyright © 2017 John Wiley & Sons, Ltd.

Generalized and synthetic regression estimators for randomized branch sampling

Treesearch

David L. R. Affleck; Timothy G. Gregoire

2015-01-01

In felled-tree studies, ratio and regression estimators are commonly used to convert more readily measured branch characteristics to dry crown mass estimates. In some cases, data from multiple trees are pooled to form these estimates. This research evaluates the utility of both tactics in the estimation of crown biomass following randomized branch sampling (...

Accounting for informatively missing data in logistic regression by means of reassessment sampling.

PubMed

Lin, Ji; Lyles, Robert H

2015-05-20

We explore the 'reassessment' design in a logistic regression setting, where a second wave of sampling is applied to recover a portion of the missing data on a binary exposure and/or outcome variable. We construct a joint likelihood function based on the original model of interest and a model for the missing data mechanism, with emphasis on non-ignorable missingness. The estimation is carried out by numerical maximization of the joint likelihood function with close approximation of the accompanying Hessian matrix, using sharable programs that take advantage of general optimization routines in standard software. We show how likelihood ratio tests can be used for model selection and how they facilitate direct hypothesis testing for whether missingness is at random. Examples and simulations are presented to demonstrate the performance of the proposed method. Copyright © 2015 John Wiley & Sons, Ltd.

An Overview of Longitudinal Data Analysis Methods for Neurological Research

PubMed Central

Locascio, Joseph J.; Atri, Alireza

2011-01-01

The purpose of this article is to provide a concise, broad and readily accessible overview of longitudinal data analysis methods, aimed to be a practical guide for clinical investigators in neurology. In general, we advise that older, traditional methods, including (1) simple regression of the dependent variable on a time measure, (2) analyzing a single summary subject level number that indexes changes for each subject and (3) a general linear model approach with a fixed-subject effect, should be reserved for quick, simple or preliminary analyses. We advocate the general use of mixed-random and fixed-effect regression models for analyses of most longitudinal clinical studies. Under restrictive situations or to provide validation, we recommend: (1) repeated-measure analysis of covariance (ANCOVA), (2) ANCOVA for two time points, (3) generalized estimating equations and (4) latent growth curve/structural equation models. PMID:22203825

Estimation of genetic effects in the presence of multicollinearity in multibreed beef cattle evaluation.

PubMed

Roso, V M; Schenkel, F S; Miller, S P; Schaeffer, L R

2005-08-01

Breed additive, dominance, and epistatic loss effects are of concern in the genetic evaluation of a multibreed population. Multiple regression equations used for fitting these effects may show a high degree of multicollinearity among predictor variables. Typically, when strong linear relationships exist, the regression coefficients have large SE and are sensitive to changes in the data file and to the addition or deletion of variables in the model. Generalized ridge regression methods were applied to obtain stable estimates of direct and maternal breed additive, dominance, and epistatic loss effects in the presence of multicollinearity among predictor variables. Preweaning weight gains of beef calves in Ontario, Canada, from 1986 to 1999 were analyzed. The genetic model included fixed direct and maternal breed additive, dominance, and epistatic loss effects, fixed environmental effects of age of the calf, contemporary group, and age of the dam x sex of the calf, random additive direct and maternal genetic effects, and random maternal permanent environment effect. The degree and the nature of the multicollinearity were identified and ridge regression methods were used as an alternative to ordinary least squares (LS). Ridge parameters were obtained using two different objective methods: 1) generalized ridge estimator of Hoerl and Kennard (R1); and 2) bootstrap in combination with cross-validation (R2). Both ridge regression methods outperformed the LS estimator with respect to mean squared error of predictions (MSEP) and variance inflation factors (VIF) computed over 100 bootstrap samples. The MSEP of R1 and R2 were similar, and they were 3% less than the MSEP of LS. The average VIF of LS, R1, and R2 were equal to 26.81, 6.10, and 4.18, respectively. Ridge regression methods were particularly effective in decreasing the multicollinearity involving predictor variables of breed additive effects. Because of a high degree of confounding between estimates of maternal dominance and direct epistatic loss effects, it was not possible to compare the relative importance of these effects with a high level of confidence. The inclusion of epistatic loss effects in the additive-dominance model did not cause noticeable reranking of sires, dams, and calves based on across-breed EBV. More precise estimates of breed effects as a result of this study may result in more stable across-breed estimated breeding values over the years.

Short-term Forecasting of the Prevalence of Trachoma: Expert Opinion, Statistical Regression, versus Transmission Models

PubMed Central

Liu, Fengchen; Porco, Travis C.; Amza, Abdou; Kadri, Boubacar; Nassirou, Baido; West, Sheila K.; Bailey, Robin L.; Keenan, Jeremy D.; Solomon, Anthony W.; Emerson, Paul M.; Gambhir, Manoj; Lietman, Thomas M.

2015-01-01

Background Trachoma programs rely on guidelines made in large part using expert opinion of what will happen with and without intervention. Large community-randomized trials offer an opportunity to actually compare forecasting methods in a masked fashion. Methods The Program for the Rapid Elimination of Trachoma trials estimated longitudinal prevalence of ocular chlamydial infection from 24 communities treated annually with mass azithromycin. Given antibiotic coverage and biannual assessments from baseline through 30 months, forecasts of the prevalence of infection in each of the 24 communities at 36 months were made by three methods: the sum of 15 experts’ opinion, statistical regression of the square-root-transformed prevalence, and a stochastic hidden Markov model of infection transmission (Susceptible-Infectious-Susceptible, or SIS model). All forecasters were masked to the 36-month results and to the other forecasts. Forecasts of the 24 communities were scored by the likelihood of the observed results and compared using Wilcoxon’s signed-rank statistic. Findings Regression and SIS hidden Markov models had significantly better likelihood than community expert opinion (p = 0.004 and p = 0.01, respectively). All forecasts scored better when perturbed to decrease Fisher’s information. Each individual expert’s forecast was poorer than the sum of experts. Interpretation Regression and SIS models performed significantly better than expert opinion, although all forecasts were overly confident. Further model refinements may score better, although would need to be tested and compared in new masked studies. Construction of guidelines that rely on forecasting future prevalence could consider use of mathematical and statistical models. PMID:26302380

Short-term Forecasting of the Prevalence of Trachoma: Expert Opinion, Statistical Regression, versus Transmission Models.

PubMed

Liu, Fengchen; Porco, Travis C; Amza, Abdou; Kadri, Boubacar; Nassirou, Baido; West, Sheila K; Bailey, Robin L; Keenan, Jeremy D; Solomon, Anthony W; Emerson, Paul M; Gambhir, Manoj; Lietman, Thomas M

2015-08-01

Trachoma programs rely on guidelines made in large part using expert opinion of what will happen with and without intervention. Large community-randomized trials offer an opportunity to actually compare forecasting methods in a masked fashion. The Program for the Rapid Elimination of Trachoma trials estimated longitudinal prevalence of ocular chlamydial infection from 24 communities treated annually with mass azithromycin. Given antibiotic coverage and biannual assessments from baseline through 30 months, forecasts of the prevalence of infection in each of the 24 communities at 36 months were made by three methods: the sum of 15 experts' opinion, statistical regression of the square-root-transformed prevalence, and a stochastic hidden Markov model of infection transmission (Susceptible-Infectious-Susceptible, or SIS model). All forecasters were masked to the 36-month results and to the other forecasts. Forecasts of the 24 communities were scored by the likelihood of the observed results and compared using Wilcoxon's signed-rank statistic. Regression and SIS hidden Markov models had significantly better likelihood than community expert opinion (p = 0.004 and p = 0.01, respectively). All forecasts scored better when perturbed to decrease Fisher's information. Each individual expert's forecast was poorer than the sum of experts. Regression and SIS models performed significantly better than expert opinion, although all forecasts were overly confident. Further model refinements may score better, although would need to be tested and compared in new masked studies. Construction of guidelines that rely on forecasting future prevalence could consider use of mathematical and statistical models. Clinicaltrials.gov NCT00792922.

Fast image interpolation via random forests.

PubMed

Huang, Jun-Jie; Siu, Wan-Chi; Liu, Tian-Rui

2015-10-01

This paper proposes a two-stage framework for fast image interpolation via random forests (FIRF). The proposed FIRF method gives high accuracy, as well as requires low computation. The underlying idea of this proposed work is to apply random forests to classify the natural image patch space into numerous subspaces and learn a linear regression model for each subspace to map the low-resolution image patch to high-resolution image patch. The FIRF framework consists of two stages. Stage 1 of the framework removes most of the ringing and aliasing artifacts in the initial bicubic interpolated image, while Stage 2 further refines the Stage 1 interpolated image. By varying the number of decision trees in the random forests and the number of stages applied, the proposed FIRF method can realize computationally scalable image interpolation. Extensive experimental results show that the proposed FIRF(3, 2) method achieves more than 0.3 dB improvement in peak signal-to-noise ratio over the state-of-the-art nonlocal autoregressive modeling (NARM) method. Moreover, the proposed FIRF(1, 1) obtains similar or better results as NARM while only takes its 0.3% computational time.

Threshold regression to accommodate a censored covariate.

PubMed

Qian, Jing; Chiou, Sy Han; Maye, Jacqueline E; Atem, Folefac; Johnson, Keith A; Betensky, Rebecca A

2018-06-22

In several common study designs, regression modeling is complicated by the presence of censored covariates. Examples of such covariates include maternal age of onset of dementia that may be right censored in an Alzheimer's amyloid imaging study of healthy subjects, metabolite measurements that are subject to limit of detection censoring in a case-control study of cardiovascular disease, and progressive biomarkers whose baseline values are of interest, but are measured post-baseline in longitudinal neuropsychological studies of Alzheimer's disease. We propose threshold regression approaches for linear regression models with a covariate that is subject to random censoring. Threshold regression methods allow for immediate testing of the significance of the effect of a censored covariate. In addition, they provide for unbiased estimation of the regression coefficient of the censored covariate. We derive the asymptotic properties of the resulting estimators under mild regularity conditions. Simulations demonstrate that the proposed estimators have good finite-sample performance, and often offer improved efficiency over existing methods. We also derive a principled method for selection of the threshold. We illustrate the approach in application to an Alzheimer's disease study that investigated brain amyloid levels in older individuals, as measured through positron emission tomography scans, as a function of maternal age of dementia onset, with adjustment for other covariates. We have developed an R package, censCov, for implementation of our method, available at CRAN. © 2018, The International Biometric Society.

Real-time model learning using Incremental Sparse Spectrum Gaussian Process Regression.

PubMed

Gijsberts, Arjan; Metta, Giorgio

2013-05-01

Novel applications in unstructured and non-stationary human environments require robots that learn from experience and adapt autonomously to changing conditions. Predictive models therefore not only need to be accurate, but should also be updated incrementally in real-time and require minimal human intervention. Incremental Sparse Spectrum Gaussian Process Regression is an algorithm that is targeted specifically for use in this context. Rather than developing a novel algorithm from the ground up, the method is based on the thoroughly studied Gaussian Process Regression algorithm, therefore ensuring a solid theoretical foundation. Non-linearity and a bounded update complexity are achieved simultaneously by means of a finite dimensional random feature mapping that approximates a kernel function. As a result, the computational cost for each update remains constant over time. Finally, algorithmic simplicity and support for automated hyperparameter optimization ensures convenience when employed in practice. Empirical validation on a number of synthetic and real-life learning problems confirms that the performance of Incremental Sparse Spectrum Gaussian Process Regression is superior with respect to the popular Locally Weighted Projection Regression, while computational requirements are found to be significantly lower. The method is therefore particularly suited for learning with real-time constraints or when computational resources are limited. Copyright © 2012 Elsevier Ltd. All rights reserved.

Artificial neural networks predict the incidence of portosplenomesenteric venous thrombosis in patients with acute pancreatitis.

PubMed

Fei, Y; Hu, J; Li, W-Q; Wang, W; Zong, G-Q

2017-03-01

Essentials Predicting the occurrence of portosplenomesenteric vein thrombosis (PSMVT) is difficult. We studied 72 patients with acute pancreatitis. Artificial neural networks modeling was more accurate than logistic regression in predicting PSMVT. Additional predictive factors may be incorporated into artificial neural networks. Objective To construct and validate artificial neural networks (ANNs) for predicting the occurrence of portosplenomesenteric venous thrombosis (PSMVT) and compare the predictive ability of the ANNs with that of logistic regression. Methods The ANNs and logistic regression modeling were constructed using simple clinical and laboratory data of 72 acute pancreatitis (AP) patients. The ANNs and logistic modeling were first trained on 48 randomly chosen patients and validated on the remaining 24 patients. The accuracy and the performance characteristics were compared between these two approaches by SPSS17.0 software. Results The training set and validation set did not differ on any of the 11 variables. After training, the back propagation network training error converged to 1 × 10 -20 , and it retained excellent pattern recognition ability. When the ANNs model was applied to the validation set, it revealed a sensitivity of 80%, specificity of 85.7%, a positive predictive value of 77.6% and negative predictive value of 90.7%. The accuracy was 83.3%. Differences could be found between ANNs modeling and logistic regression modeling in these parameters (10.0% [95% CI, -14.3 to 34.3%], 14.3% [95% CI, -8.6 to 37.2%], 15.7% [95% CI, -9.9 to 41.3%], 11.8% [95% CI, -8.2 to 31.8%], 22.6% [95% CI, -1.9 to 47.1%], respectively). When ANNs modeling was used to identify PSMVT, the area under receiver operating characteristic curve was 0.849 (95% CI, 0.807-0.901), which demonstrated better overall properties than logistic regression modeling (AUC = 0.716) (95% CI, 0.679-0.761). Conclusions ANNs modeling was a more accurate tool than logistic regression in predicting the occurrence of PSMVT following AP. More clinical factors or biomarkers may be incorporated into ANNs modeling to improve its predictive ability. © 2016 International Society on Thrombosis and Haemostasis.

Comparing cluster-level dynamic treatment regimens using sequential, multiple assignment, randomized trials: Regression estimation and sample size considerations.

PubMed

NeCamp, Timothy; Kilbourne, Amy; Almirall, Daniel

2017-08-01

Cluster-level dynamic treatment regimens can be used to guide sequential treatment decision-making at the cluster level in order to improve outcomes at the individual or patient-level. In a cluster-level dynamic treatment regimen, the treatment is potentially adapted and re-adapted over time based on changes in the cluster that could be impacted by prior intervention, including aggregate measures of the individuals or patients that compose it. Cluster-randomized sequential multiple assignment randomized trials can be used to answer multiple open questions preventing scientists from developing high-quality cluster-level dynamic treatment regimens. In a cluster-randomized sequential multiple assignment randomized trial, sequential randomizations occur at the cluster level and outcomes are observed at the individual level. This manuscript makes two contributions to the design and analysis of cluster-randomized sequential multiple assignment randomized trials. First, a weighted least squares regression approach is proposed for comparing the mean of a patient-level outcome between the cluster-level dynamic treatment regimens embedded in a sequential multiple assignment randomized trial. The regression approach facilitates the use of baseline covariates which is often critical in the analysis of cluster-level trials. Second, sample size calculators are derived for two common cluster-randomized sequential multiple assignment randomized trial designs for use when the primary aim is a between-dynamic treatment regimen comparison of the mean of a continuous patient-level outcome. The methods are motivated by the Adaptive Implementation of Effective Programs Trial which is, to our knowledge, the first-ever cluster-randomized sequential multiple assignment randomized trial in psychiatry.

Multicenter Comparison of Machine Learning Methods and Conventional Regression for Predicting Clinical Deterioration on the Wards.

PubMed

Churpek, Matthew M; Yuen, Trevor C; Winslow, Christopher; Meltzer, David O; Kattan, Michael W; Edelson, Dana P

2016-02-01

Machine learning methods are flexible prediction algorithms that may be more accurate than conventional regression. We compared the accuracy of different techniques for detecting clinical deterioration on the wards in a large, multicenter database. Observational cohort study. Five hospitals, from November 2008 until January 2013. Hospitalized ward patients None Demographic variables, laboratory values, and vital signs were utilized in a discrete-time survival analysis framework to predict the combined outcome of cardiac arrest, intensive care unit transfer, or death. Two logistic regression models (one using linear predictor terms and a second utilizing restricted cubic splines) were compared to several different machine learning methods. The models were derived in the first 60% of the data by date and then validated in the next 40%. For model derivation, each event time window was matched to a non-event window. All models were compared to each other and to the Modified Early Warning score, a commonly cited early warning score, using the area under the receiver operating characteristic curve (AUC). A total of 269,999 patients were admitted, and 424 cardiac arrests, 13,188 intensive care unit transfers, and 2,840 deaths occurred in the study. In the validation dataset, the random forest model was the most accurate model (AUC, 0.80 [95% CI, 0.80-0.80]). The logistic regression model with spline predictors was more accurate than the model utilizing linear predictors (AUC, 0.77 vs 0.74; p < 0.01), and all models were more accurate than the MEWS (AUC, 0.70 [95% CI, 0.70-0.70]). In this multicenter study, we found that several machine learning methods more accurately predicted clinical deterioration than logistic regression. Use of detection algorithms derived from these techniques may result in improved identification of critically ill patients on the wards.

Producing landslide susceptibility maps by utilizing machine learning methods. The case of Finikas catchment basin, North Peloponnese, Greece.

NASA Astrophysics Data System (ADS)

Tsangaratos, Paraskevas; Ilia, Ioanna; Loupasakis, Constantinos; Papadakis, Michalis; Karimalis, Antonios

2017-04-01

The main objective of the present study was to apply two machine learning methods for the production of a landslide susceptibility map in the Finikas catchment basin, located in North Peloponnese, Greece and to compare their results. Specifically, Logistic Regression and Random Forest were utilized, based on a database of 40 sites classified into two categories, non-landslide and landslide areas that were separated into a training dataset (70% of the total data) and a validation dataset (remaining 30%). The identification of the areas was established by analyzing airborne imagery, extensive field investigation and the examination of previous research studies. Six landslide related variables were analyzed, namely: lithology, elevation, slope, aspect, distance to rivers and distance to faults. Within the Finikas catchment basin most of the reported landslides were located along the road network and within the residential complexes, classified as rotational and translational slides, and rockfalls, mainly caused due to the physical conditions and the general geotechnical behavior of the geological formation that cover the area. Each landslide susceptibility map was reclassified by applying the Geometric Interval classification technique into five classes, namely: very low susceptibility, low susceptibility, moderate susceptibility, high susceptibility, and very high susceptibility. The comparison and validation of the outcomes of each model were achieved using statistical evaluation measures, the receiving operating characteristic and the area under the success and predictive rate curves. The computation process was carried out using RStudio an integrated development environment for R language and ArcGIS 10.1 for compiling the data and producing the landslide susceptibility maps. From the outcomes of the Logistic Regression analysis it was induced that the highest b coefficient is allocated to lithology and slope, which was 2.8423 and 1.5841, respectively. From the estimation of the mean decrease in Gini coefficient performed during the application of Random Forest and the mean decrease in accuracy the most important variable is slope followed by lithology, aspect, elevation, distance from river network, and distance from faults, while the most used variables during the training phase were the variable aspect (21.45%), slope (20.53%) and lithology (19.84%). The outcomes of the analysis are consistent with previous studies concerning the area of research, which have indicated the high influence of lithology and slope in the manifestation of landslides. High percentage of landslide occurrence has been observed in Plio-Pleistocene sediments, flysch formations, and Cretaceous limestone. Also the presences of landslides have been associated with the degree of weathering and fragmentation, the orientation of the discontinuities surfaces and the intense morphological relief. The most accurate model was Random Forest which identified correctly 92.00% of the instances during the training phase, followed by the Logistic Regression 89.00%. The same pattern of accuracy was calculated during the validation phase, in which the Random Forest achieved a classification accuracy of 93.00%, while the Logistic Regression model achieved an accuracy of 91.00%. In conclusion, the outcomes of the study could be a useful cartographic product to local authorities and government agencies during the implementation of successful decision-making and land use planning strategies. Keywords: Landslide Susceptibility, Logistic Regression, Random Forest, GIS, Greece.

Inner and outer coronary vessel wall segmentation from CCTA using an active contour model with machine learning-based 3D voxel context-aware image force

NASA Astrophysics Data System (ADS)

Sivalingam, Udhayaraj; Wels, Michael; Rempfler, Markus; Grosskopf, Stefan; Suehling, Michael; Menze, Bjoern H.

2016-03-01

In this paper, we present a fully automated approach to coronary vessel segmentation, which involves calcification or soft plaque delineation in addition to accurate lumen delineation, from 3D Cardiac Computed Tomography Angiography data. Adequately virtualizing the coronary lumen plays a crucial role for simulating blood ow by means of fluid dynamics while additionally identifying the outer vessel wall in the case of arteriosclerosis is a prerequisite for further plaque compartment analysis. Our method is a hybrid approach complementing Active Contour Model-based segmentation with an external image force that relies on a Random Forest Regression model generated off-line. The regression model provides a strong estimate of the distance to the true vessel surface for every surface candidate point taking into account 3D wavelet-encoded contextual image features, which are aligned with the current surface hypothesis. The associated external image force is integrated in the objective function of the active contour model, such that the overall segmentation approach benefits from the advantages associated with snakes and from the ones associated with machine learning-based regression alike. This yields an integrated approach achieving competitive results on a publicly available benchmark data collection (Rotterdam segmentation challenge).

Development and validation of a mortality risk model for pediatric sepsis.

PubMed

Chen, Mengshi; Lu, Xiulan; Hu, Li; Liu, Pingping; Zhao, Wenjiao; Yan, Haipeng; Tang, Liang; Zhu, Yimin; Xiao, Zhenghui; Chen, Lizhang; Tan, Hongzhuan

2017-05-01

Pediatric sepsis is a burdensome public health problem. Assessing the mortality risk of pediatric sepsis patients, offering effective treatment guidance, and improving prognosis to reduce mortality rates, are crucial.We extracted data derived from electronic medical records of pediatric sepsis patients that were collected during the first 24 hours after admission to the pediatric intensive care unit (PICU) of the Hunan Children's hospital from January 2012 to June 2014. A total of 788 children were randomly divided into a training (592, 75%) and validation group (196, 25%). The risk factors for mortality among these patients were identified by conducting multivariate logistic regression in the training group. Based on the established logistic regression equation, the logit probabilities for all patients (in both groups) were calculated to verify the model's internal and external validities.According to the training group, 6 variables (brain natriuretic peptide, albumin, total bilirubin, D-dimer, lactate levels, and mechanical ventilation in 24 hours) were included in the final logistic regression model. The areas under the curves of the model were 0.854 (0.826, 0.881) and 0.844 (0.816, 0.873) in the training and validation groups, respectively.The Mortality Risk Model for Pediatric Sepsis we established in this study showed acceptable accuracy to predict the mortality risk in pediatric sepsis patients.

Development and validation of a mortality risk model for pediatric sepsis

PubMed Central

Chen, Mengshi; Lu, Xiulan; Hu, Li; Liu, Pingping; Zhao, Wenjiao; Yan, Haipeng; Tang, Liang; Zhu, Yimin; Xiao, Zhenghui; Chen, Lizhang; Tan, Hongzhuan

2017-01-01

Abstract Pediatric sepsis is a burdensome public health problem. Assessing the mortality risk of pediatric sepsis patients, offering effective treatment guidance, and improving prognosis to reduce mortality rates, are crucial. We extracted data derived from electronic medical records of pediatric sepsis patients that were collected during the first 24 hours after admission to the pediatric intensive care unit (PICU) of the Hunan Children's hospital from January 2012 to June 2014. A total of 788 children were randomly divided into a training (592, 75%) and validation group (196, 25%). The risk factors for mortality among these patients were identified by conducting multivariate logistic regression in the training group. Based on the established logistic regression equation, the logit probabilities for all patients (in both groups) were calculated to verify the model's internal and external validities. According to the training group, 6 variables (brain natriuretic peptide, albumin, total bilirubin, D-dimer, lactate levels, and mechanical ventilation in 24 hours) were included in the final logistic regression model. The areas under the curves of the model were 0.854 (0.826, 0.881) and 0.844 (0.816, 0.873) in the training and validation groups, respectively. The Mortality Risk Model for Pediatric Sepsis we established in this study showed acceptable accuracy to predict the mortality risk in pediatric sepsis patients. PMID:28514310

Solvency supervision based on a total balance sheet approach

NASA Astrophysics Data System (ADS)

Pitselis, Georgios

2009-11-01

In this paper we investigate the adequacy of the own funds a company requires in order to remain healthy and avoid insolvency. Two methods are applied here; the quantile regression method and the method of mixed effects models. Quantile regression is capable of providing a more complete statistical analysis of the stochastic relationship among random variables than least squares estimation. The estimated mixed effects line can be considered as an internal industry equation (norm), which explains a systematic relation between a dependent variable (such as own funds) with independent variables (e.g. financial characteristics, such as assets, provisions, etc.). The above two methods are implemented with two data sets.

Classification of suicide attempters in schizophrenia using sociocultural and clinical features: A machine learning approach.

PubMed

Hettige, Nuwan C; Nguyen, Thai Binh; Yuan, Chen; Rajakulendran, Thanara; Baddour, Jermeen; Bhagwat, Nikhil; Bani-Fatemi, Ali; Voineskos, Aristotle N; Mallar Chakravarty, M; De Luca, Vincenzo

2017-07-01

Suicide is a major concern for those afflicted by schizophrenia. Identifying patients at the highest risk for future suicide attempts remains a complex problem for psychiatric interventions. Machine learning models allow for the integration of many risk factors in order to build an algorithm that predicts which patients are likely to attempt suicide. Currently it is unclear how to integrate previously identified risk factors into a clinically relevant predictive tool to estimate the probability of a patient with schizophrenia for attempting suicide. We conducted a cross-sectional assessment on a sample of 345 participants diagnosed with schizophrenia spectrum disorders. Suicide attempters and non-attempters were clearly identified using the Columbia Suicide Severity Rating Scale (C-SSRS) and the Beck Suicide Ideation Scale (BSS). We developed four classification algorithms using a regularized regression, random forest, elastic net and support vector machine models with sociocultural and clinical variables as features to train the models. All classification models performed similarly in identifying suicide attempters and non-attempters. Our regularized logistic regression model demonstrated an accuracy of 67% and an area under the curve (AUC) of 0.71, while the random forest model demonstrated 66% accuracy and an AUC of 0.67. Support vector classifier (SVC) model demonstrated an accuracy of 67% and an AUC of 0.70, and the elastic net model demonstrated and accuracy of 65% and an AUC of 0.71. Machine learning algorithms offer a relatively successful method for incorporating many clinical features to predict individuals at risk for future suicide attempts. Increased performance of these models using clinically relevant variables offers the potential to facilitate early treatment and intervention to prevent future suicide attempts. Copyright © 2017 Elsevier Inc. All rights reserved.

Taxi-Out Time Prediction for Departures at Charlotte Airport Using Machine Learning Techniques

NASA Technical Reports Server (NTRS)

Lee, Hanbong; Malik, Waqar; Jung, Yoon C.

2016-01-01

Predicting the taxi-out times of departures accurately is important for improving airport efficiency and takeoff time predictability. In this paper, we attempt to apply machine learning techniques to actual traffic data at Charlotte Douglas International Airport for taxi-out time prediction. To find the key factors affecting aircraft taxi times, surface surveillance data is first analyzed. From this data analysis, several variables, including terminal concourse, spot, runway, departure fix and weight class, are selected for taxi time prediction. Then, various machine learning methods such as linear regression, support vector machines, k-nearest neighbors, random forest, and neural networks model are applied to actual flight data. Different traffic flow and weather conditions at Charlotte airport are also taken into account for more accurate prediction. The taxi-out time prediction results show that linear regression and random forest techniques can provide the most accurate prediction in terms of root-mean-square errors. We also discuss the operational complexity and uncertainties that make it difficult to predict the taxi times accurately.

Quantitative monitoring of sucrose, reducing sugar and total sugar dynamics for phenotyping of water-deficit stress tolerance in rice through spectroscopy and chemometrics

NASA Astrophysics Data System (ADS)

Das, Bappa; Sahoo, Rabi N.; Pargal, Sourabh; Krishna, Gopal; Verma, Rakesh; Chinnusamy, Viswanathan; Sehgal, Vinay K.; Gupta, Vinod K.; Dash, Sushanta K.; Swain, Padmini

2018-03-01

In the present investigation, the changes in sucrose, reducing and total sugar content due to water-deficit stress in rice leaves were modeled using visible, near infrared (VNIR) and shortwave infrared (SWIR) spectroscopy. The objectives of the study were to identify the best vegetation indices and suitable multivariate technique based on precise analysis of hyperspectral data (350 to 2500 nm) and sucrose, reducing sugar and total sugar content measured at different stress levels from 16 different rice genotypes. Spectral data analysis was done to identify suitable spectral indices and models for sucrose estimation. Novel spectral indices in near infrared (NIR) range viz. ratio spectral index (RSI) and normalised difference spectral indices (NDSI) sensitive to sucrose, reducing sugar and total sugar content were identified which were subsequently calibrated and validated. The RSI and NDSI models had R2 values of 0.65, 0.71 and 0.67; RPD values of 1.68, 1.95 and 1.66 for sucrose, reducing sugar and total sugar, respectively for validation dataset. Different multivariate spectral models such as artificial neural network (ANN), multivariate adaptive regression splines (MARS), multiple linear regression (MLR), partial least square regression (PLSR), random forest regression (RFR) and support vector machine regression (SVMR) were also evaluated. The best performing multivariate models for sucrose, reducing sugars and total sugars were found to be, MARS, ANN and MARS, respectively with respect to RPD values of 2.08, 2.44, and 1.93. Results indicated that VNIR and SWIR spectroscopy combined with multivariate calibration can be used as a reliable alternative to conventional methods for measurement of sucrose, reducing sugars and total sugars of rice under water-deficit stress as this technique is fast, economic, and noninvasive.

A Ranking Approach to Genomic Selection.

PubMed

Blondel, Mathieu; Onogi, Akio; Iwata, Hiroyoshi; Ueda, Naonori

2015-01-01

Genomic selection (GS) is a recent selective breeding method which uses predictive models based on whole-genome molecular markers. Until now, existing studies formulated GS as the problem of modeling an individual's breeding value for a particular trait of interest, i.e., as a regression problem. To assess predictive accuracy of the model, the Pearson correlation between observed and predicted trait values was used. In this paper, we propose to formulate GS as the problem of ranking individuals according to their breeding value. Our proposed framework allows us to employ machine learning methods for ranking which had previously not been considered in the GS literature. To assess ranking accuracy of a model, we introduce a new measure originating from the information retrieval literature called normalized discounted cumulative gain (NDCG). NDCG rewards more strongly models which assign a high rank to individuals with high breeding value. Therefore, NDCG reflects a prerequisite objective in selective breeding: accurate selection of individuals with high breeding value. We conducted a comparison of 10 existing regression methods and 3 new ranking methods on 6 datasets, consisting of 4 plant species and 25 traits. Our experimental results suggest that tree-based ensemble methods including McRank, Random Forests and Gradient Boosting Regression Trees achieve excellent ranking accuracy. RKHS regression and RankSVM also achieve good accuracy when used with an RBF kernel. Traditional regression methods such as Bayesian lasso, wBSR and BayesC were found less suitable for ranking. Pearson correlation was found to correlate poorly with NDCG. Our study suggests two important messages. First, ranking methods are a promising research direction in GS. Second, NDCG can be a useful evaluation measure for GS.

Model selection with multiple regression on distance matrices leads to incorrect inferences.

PubMed

Franckowiak, Ryan P; Panasci, Michael; Jarvis, Karl J; Acuña-Rodriguez, Ian S; Landguth, Erin L; Fortin, Marie-Josée; Wagner, Helene H

2017-01-01

In landscape genetics, model selection procedures based on Information Theoretic and Bayesian principles have been used with multiple regression on distance matrices (MRM) to test the relationship between multiple vectors of pairwise genetic, geographic, and environmental distance. Using Monte Carlo simulations, we examined the ability of model selection criteria based on Akaike's information criterion (AIC), its small-sample correction (AICc), and the Bayesian information criterion (BIC) to reliably rank candidate models when applied with MRM while varying the sample size. The results showed a serious problem: all three criteria exhibit a systematic bias toward selecting unnecessarily complex models containing spurious random variables and erroneously suggest a high level of support for the incorrectly ranked best model. These problems effectively increased with increasing sample size. The failure of AIC, AICc, and BIC was likely driven by the inflated sample size and different sum-of-squares partitioned by MRM, and the resulting effect on delta values. Based on these findings, we strongly discourage the continued application of AIC, AICc, and BIC for model selection with MRM.

Robustness of intra urban land-use regression models for ultrafine particles and black carbon based on mobile monitoring.

PubMed

Kerckhoffs, Jules; Hoek, Gerard; Vlaanderen, Jelle; van Nunen, Erik; Messier, Kyle; Brunekreef, Bert; Gulliver, John; Vermeulen, Roel

2017-11-01

Land-use regression (LUR) models for ultrafine particles (UFP) and Black Carbon (BC) in urban areas have been developed using short-term stationary monitoring or mobile platforms in order to capture the high variability of these pollutants. However, little is known about the comparability of predictions of mobile and short-term stationary models and especially the validity of these models for assessing residential exposures and the robustness of model predictions developed in different campaigns. We used an electric car to collect mobile measurements (n = 5236 unique road segments) and short-term stationary measurements (3 × 30min, n = 240) of UFP and BC in three Dutch cities (Amsterdam, Utrecht, Maastricht) in 2014-2015. Predictions of LUR models based on mobile measurements were compared to (i) measured concentrations at the short-term stationary sites, (ii) LUR model predictions based on short-term stationary measurements at 1500 random addresses in the three cities, (iii) externally obtained home outdoor measurements (3 × 24h samples; n = 42) and (iv) predictions of a LUR model developed based upon a 2013 mobile campaign in two cities (Amsterdam, Rotterdam). Despite the poor model R 2 of 15%, the ability of mobile UFP models to predict measurements with longer averaging time increased substantially from 36% for short-term stationary measurements to 57% for home outdoor measurements. In contrast, the mobile BC model only predicted 14% of the variation in the short-term stationary sites and also 14% of the home outdoor sites. Models based upon mobile and short-term stationary monitoring provided fairly high correlated predictions of UFP concentrations at 1500 randomly selected addresses in the three Dutch cities (R 2 = 0.64). We found higher UFP predictions (of about 30%) based on mobile models opposed to short-term model predictions and home outdoor measurements with no clear geospatial patterns. The mobile model for UFP was stable over different settings as the model predicted concentration levels highly correlated to predictions made by a previously developed LUR model with another spatial extent and in a different year at the 1500 random addresses (R 2 = 0.80). In conclusion, mobile monitoring provided robust LUR models for UFP, valid to use in epidemiological studies. Copyright © 2017 Elsevier Inc. All rights reserved.

Effective search for stable segregation configurations at grain boundaries with data-mining techniques

NASA Astrophysics Data System (ADS)

Kiyohara, Shin; Mizoguchi, Teruyasu

2018-03-01

Grain boundary segregation of dopants plays a crucial role in materials properties. To investigate the dopant segregation behavior at the grain boundary, an enormous number of combinations have to be considered in the segregation of multiple dopants at the complex grain boundary structures. Here, two data mining techniques, the random-forests regression and the genetic algorithm, were applied to determine stable segregation sites at grain boundaries efficiently. Using the random-forests method, a predictive model was constructed from 2% of the segregation configurations and it has been shown that this model could determine the stable segregation configurations. Furthermore, the genetic algorithm also successfully determined the most stable segregation configuration with great efficiency. We demonstrate that these approaches are quite effective to investigate the dopant segregation behaviors at grain boundaries.

An Introduction to Recursive Partitioning: Rationale, Application, and Characteristics of Classification and Regression Trees, Bagging, and Random Forests

ERIC Educational Resources Information Center

Strobl, Carolin; Malley, James; Tutz, Gerhard

2009-01-01

Recursive partitioning methods have become popular and widely used tools for nonparametric regression and classification in many scientific fields. Especially random forests, which can deal with large numbers of predictor variables even in the presence of complex interactions, have been applied successfully in genetics, clinical medicine, and…

Subpixel urban land cover estimation: comparing cubist, random forests, and support vector regression

Treesearch

Jeffrey T. Walton

2008-01-01

Three machine learning subpixel estimation methods (Cubist, Random Forests, and support vector regression) were applied to estimate urban cover. Urban forest canopy cover and impervious surface cover were estimated from Landsat-7 ETM+ imagery using a higher resolution cover map resampled to 30 m as training and reference data. Three different band combinations (...

An Empirical Comparison of Randomized Control Trials and Regression Discontinuity Estimations

ERIC Educational Resources Information Center

Barrera-Osorio, Felipe; Filmer, Deon; McIntyre, Joe

2014-01-01

Randomized controlled trials (RCTs) and regression discontinuity (RD) studies both provide estimates of causal effects. A major difference between the two is that RD only estimates local average treatment effects (LATE) near the cutoff point of the forcing variable. This has been cited as a drawback to RD designs (Cook & Wong, 2008).…

A test of geographic assignment using isotope tracers in feathers of known origin

USGS Publications Warehouse

Wunder, Michael B.; Kester, C.L.; Knopf, F.L.; Rye, R.O.

2005-01-01

We used feathers of known origin collected from across the breeding range of a migratory shorebird to test the use of isotope tracers for assigning breeding origins. We analyzed δD, δ13C, and δ15N in feathers from 75 mountain plover (Charadrius montanus) chicks sampled in 2001 and from 119 chicks sampled in 2002. We estimated parameters for continuous-response inverse regression models and for discrete-response Bayesian probability models from data for each year independently. We evaluated model predictions with both the training data and by using the alternate year as an independent test dataset. Our results provide weak support for modeling latitude and isotope values as monotonic functions of one another, especially when data are pooled over known sources of variation such as sample year or location. We were unable to make even qualitative statements, such as north versus south, about the likely origin of birds using both δD and δ13C in inverse regression models; results were no better than random assignment. Probability models provided better results and a more natural framework for the problem. Correct assignment rates were highest when considering all three isotopes in the probability framework, but the use of even a single isotope was better than random assignment. The method appears relatively robust to temporal effects and is most sensitive to the isotope discrimination gradients over which samples are taken. We offer that the problem of using isotope tracers to infer geographic origin is best framed as one of assignment, rather than prediction.

Sampling Strategies for Evaluating the Rate of Adventitious Transgene Presence in Non-Genetically Modified Crop Fields.

PubMed

Makowski, David; Bancal, Rémi; Bensadoun, Arnaud; Monod, Hervé; Messéan, Antoine

2017-09-01

According to E.U. regulations, the maximum allowable rate of adventitious transgene presence in non-genetically modified (GM) crops is 0.9%. We compared four sampling methods for the detection of transgenic material in agricultural non-GM maize fields: random sampling, stratified sampling, random sampling + ratio reweighting, random sampling + regression reweighting. Random sampling involves simply sampling maize grains from different locations selected at random from the field concerned. The stratified and reweighting sampling methods make use of an auxiliary variable corresponding to the output of a gene-flow model (a zero-inflated Poisson model) simulating cross-pollination as a function of wind speed, wind direction, and distance to the closest GM maize field. With the stratified sampling method, an auxiliary variable is used to define several strata with contrasting transgene presence rates, and grains are then sampled at random from each stratum. With the two methods involving reweighting, grains are first sampled at random from various locations within the field, and the observations are then reweighted according to the auxiliary variable. Data collected from three maize fields were used to compare the four sampling methods, and the results were used to determine the extent to which transgene presence rate estimation was improved by the use of stratified and reweighting sampling methods. We found that transgene rate estimates were more accurate and that substantially smaller samples could be used with sampling strategies based on an auxiliary variable derived from a gene-flow model. © 2017 Society for Risk Analysis.

Nonparametric estimation and testing of fixed effects panel data models

PubMed Central

Henderson, Daniel J.; Carroll, Raymond J.; Li, Qi

2009-01-01

In this paper we consider the problem of estimating nonparametric panel data models with fixed effects. We introduce an iterative nonparametric kernel estimator. We also extend the estimation method to the case of a semiparametric partially linear fixed effects model. To determine whether a parametric, semiparametric or nonparametric model is appropriate, we propose test statistics to test between the three alternatives in practice. We further propose a test statistic for testing the null hypothesis of random effects against fixed effects in a nonparametric panel data regression model. Simulations are used to examine the finite sample performance of the proposed estimators and the test statistics. PMID:19444335

Modeling Linguistic Variables With Regression Models: Addressing Non-Gaussian Distributions, Non-independent Observations, and Non-linear Predictors With Random Effects and Generalized Additive Models for Location, Scale, and Shape

PubMed Central

Coupé, Christophe

2018-01-01

As statistical approaches are getting increasingly used in linguistics, attention must be paid to the choice of methods and algorithms used. This is especially true since they require assumptions to be satisfied to provide valid results, and because scientific articles still often fall short of reporting whether such assumptions are met. Progress is being, however, made in various directions, one of them being the introduction of techniques able to model data that cannot be properly analyzed with simpler linear regression models. We report recent advances in statistical modeling in linguistics. We first describe linear mixed-effects regression models (LMM), which address grouping of observations, and generalized linear mixed-effects models (GLMM), which offer a family of distributions for the dependent variable. Generalized additive models (GAM) are then introduced, which allow modeling non-linear parametric or non-parametric relationships between the dependent variable and the predictors. We then highlight the possibilities offered by generalized additive models for location, scale, and shape (GAMLSS). We explain how they make it possible to go beyond common distributions, such as Gaussian or Poisson, and offer the appropriate inferential framework to account for ‘difficult’ variables such as count data with strong overdispersion. We also demonstrate how they offer interesting perspectives on data when not only the mean of the dependent variable is modeled, but also its variance, skewness, and kurtosis. As an illustration, the case of phonemic inventory size is analyzed throughout the article. For over 1,500 languages, we consider as predictors the number of speakers, the distance from Africa, an estimation of the intensity of language contact, and linguistic relationships. We discuss the use of random effects to account for genealogical relationships, the choice of appropriate distributions to model count data, and non-linear relationships. Relying on GAMLSS, we assess a range of candidate distributions, including the Sichel, Delaporte, Box-Cox Green and Cole, and Box-Cox t distributions. We find that the Box-Cox t distribution, with appropriate modeling of its parameters, best fits the conditional distribution of phonemic inventory size. We finally discuss the specificities of phoneme counts, weak effects, and how GAMLSS should be considered for other linguistic variables. PMID:29713298

Modeling Linguistic Variables With Regression Models: Addressing Non-Gaussian Distributions, Non-independent Observations, and Non-linear Predictors With Random Effects and Generalized Additive Models for Location, Scale, and Shape.

PubMed

Coupé, Christophe

2018-01-01

As statistical approaches are getting increasingly used in linguistics, attention must be paid to the choice of methods and algorithms used. This is especially true since they require assumptions to be satisfied to provide valid results, and because scientific articles still often fall short of reporting whether such assumptions are met. Progress is being, however, made in various directions, one of them being the introduction of techniques able to model data that cannot be properly analyzed with simpler linear regression models. We report recent advances in statistical modeling in linguistics. We first describe linear mixed-effects regression models (LMM), which address grouping of observations, and generalized linear mixed-effects models (GLMM), which offer a family of distributions for the dependent variable. Generalized additive models (GAM) are then introduced, which allow modeling non-linear parametric or non-parametric relationships between the dependent variable and the predictors. We then highlight the possibilities offered by generalized additive models for location, scale, and shape (GAMLSS). We explain how they make it possible to go beyond common distributions, such as Gaussian or Poisson, and offer the appropriate inferential framework to account for 'difficult' variables such as count data with strong overdispersion. We also demonstrate how they offer interesting perspectives on data when not only the mean of the dependent variable is modeled, but also its variance, skewness, and kurtosis. As an illustration, the case of phonemic inventory size is analyzed throughout the article. For over 1,500 languages, we consider as predictors the number of speakers, the distance from Africa, an estimation of the intensity of language contact, and linguistic relationships. We discuss the use of random effects to account for genealogical relationships, the choice of appropriate distributions to model count data, and non-linear relationships. Relying on GAMLSS, we assess a range of candidate distributions, including the Sichel, Delaporte, Box-Cox Green and Cole, and Box-Cox t distributions. We find that the Box-Cox t distribution, with appropriate modeling of its parameters, best fits the conditional distribution of phonemic inventory size. We finally discuss the specificities of phoneme counts, weak effects, and how GAMLSS should be considered for other linguistic variables.

Improving riverine constituent concentration and flux estimation by accounting for antecedent discharge conditions

NASA Astrophysics Data System (ADS)

Zhang, Qian; Ball, William P.

2017-04-01

Regression-based approaches are often employed to estimate riverine constituent concentrations and fluxes based on typically sparse concentration observations. One such approach is the recently developed WRTDS ("Weighted Regressions on Time, Discharge, and Season") method, which has been shown to provide more accurate estimates than prior approaches in a wide range of applications. Centered on WRTDS, this work was aimed at developing improved models for constituent concentration and flux estimation by accounting for antecedent discharge conditions. Twelve modified models were developed and tested, each of which contains one additional flow variable to represent antecedent conditions and which can be directly derived from the daily discharge record. High-resolution (∼daily) data at nine diverse monitoring sites were used to evaluate the relative merits of the models for estimation of six constituents - chloride (Cl), nitrate-plus-nitrite (NOx), total Kjeldahl nitrogen (TKN), total phosphorus (TP), soluble reactive phosphorus (SRP), and suspended sediment (SS). For each site-constituent combination, 30 concentration subsets were generated from the original data through Monte Carlo subsampling and then used to evaluate model performance. For the subsampling, three sampling strategies were adopted: (A) 1 random sample each month (12/year), (B) 12 random monthly samples plus additional 8 random samples per year (20/year), and (C) flow-stratified sampling with 12 regular (non-storm) and 8 storm samples per year (20/year). Results reveal that estimation performance varies with both model choice and sampling strategy. In terms of model choice, the modified models show general improvement over the original model under all three sampling strategies. Major improvements were achieved for NOx by the long-term flow-anomaly model and for Cl by the ADF (average discounted flow) model and the short-term flow-anomaly model. Moderate improvements were achieved for SS, TP, and TKN by the ADF model. By contrast, no such achievement was achieved for SRP by any proposed model. In terms of sampling strategy, performance of all models (including the original) was generally best using strategy C and worst using strategy A, and especially so for SS, TP, and SRP, confirming the value of routinely collecting stormflow samples. Overall, this work provides a comprehensive set of statistical evidence for supporting the incorporation of antecedent discharge conditions into the WRTDS model for estimation of constituent concentration and flux, thereby combining the advantages of two recent developments in water quality modeling.

Gender, Alcohol Consumption Patterns, and Engagement in Sexually Intimate Behaviors among Adolescents and Young Adults in Nha Trang, Viet Nam

ERIC Educational Resources Information Center

Kaljee, Linda M.; Green, Mackenzie S.; Zhan, Min; Riel, Rosemary; Lerdboon, Porntip; Lostutter, Ty W.; Tho, Le Huu; Luong, Vo Van; Minh, Truong Tan

2011-01-01

A randomly selected cross-sectional survey was conducted with 880 youth (16 to 24 years) in Nha Trang City to assess relationships between alcohol consumption and sexual behaviors. A timeline followback method was employed. Chi-square, generalized logit modeling and logistic regression analyses were performed. Of the sample, 78.2% male and 56.1%…

Accounting for Heterogeneity in Relative Treatment Effects for Use in Cost-Effectiveness Models and Value-of-Information Analyses

PubMed Central

Soares, Marta O.; Palmer, Stephen; Ades, Anthony E.; Harrison, David; Shankar-Hari, Manu; Rowan, Kathy M.

2015-01-01

Cost-effectiveness analysis (CEA) models are routinely used to inform health care policy. Key model inputs include relative effectiveness of competing treatments, typically informed by meta-analysis. Heterogeneity is ubiquitous in meta-analysis, and random effects models are usually used when there is variability in effects across studies. In the absence of observed treatment effect modifiers, various summaries from the random effects distribution (random effects mean, predictive distribution, random effects distribution, or study-specific estimate [shrunken or independent of other studies]) can be used depending on the relationship between the setting for the decision (population characteristics, treatment definitions, and other contextual factors) and the included studies. If covariates have been measured that could potentially explain the heterogeneity, then these can be included in a meta-regression model. We describe how covariates can be included in a network meta-analysis model and how the output from such an analysis can be used in a CEA model. We outline a model selection procedure to help choose between competing models and stress the importance of clinical input. We illustrate the approach with a health technology assessment of intravenous immunoglobulin for the management of adult patients with severe sepsis in an intensive care setting, which exemplifies how risk of bias information can be incorporated into CEA models. We show that the results of the CEA and value-of-information analyses are sensitive to the model and highlight the importance of sensitivity analyses when conducting CEA in the presence of heterogeneity. The methods presented extend naturally to heterogeneity in other model inputs, such as baseline risk. PMID:25712447

Accounting for Heterogeneity in Relative Treatment Effects for Use in Cost-Effectiveness Models and Value-of-Information Analyses.

PubMed

Welton, Nicky J; Soares, Marta O; Palmer, Stephen; Ades, Anthony E; Harrison, David; Shankar-Hari, Manu; Rowan, Kathy M

2015-07-01

Cost-effectiveness analysis (CEA) models are routinely used to inform health care policy. Key model inputs include relative effectiveness of competing treatments, typically informed by meta-analysis. Heterogeneity is ubiquitous in meta-analysis, and random effects models are usually used when there is variability in effects across studies. In the absence of observed treatment effect modifiers, various summaries from the random effects distribution (random effects mean, predictive distribution, random effects distribution, or study-specific estimate [shrunken or independent of other studies]) can be used depending on the relationship between the setting for the decision (population characteristics, treatment definitions, and other contextual factors) and the included studies. If covariates have been measured that could potentially explain the heterogeneity, then these can be included in a meta-regression model. We describe how covariates can be included in a network meta-analysis model and how the output from such an analysis can be used in a CEA model. We outline a model selection procedure to help choose between competing models and stress the importance of clinical input. We illustrate the approach with a health technology assessment of intravenous immunoglobulin for the management of adult patients with severe sepsis in an intensive care setting, which exemplifies how risk of bias information can be incorporated into CEA models. We show that the results of the CEA and value-of-information analyses are sensitive to the model and highlight the importance of sensitivity analyses when conducting CEA in the presence of heterogeneity. The methods presented extend naturally to heterogeneity in other model inputs, such as baseline risk. © The Author(s) 2015.

Genetic Analysis of Milk Yield in First-Lactation Holstein Friesian in Ethiopia: A Lactation Average vs Random Regression Test-Day Model Analysis

PubMed Central

Meseret, S.; Tamir, B.; Gebreyohannes, G.; Lidauer, M.; Negussie, E.

2015-01-01

The development of effective genetic evaluations and selection of sires requires accurate estimates of genetic parameters for all economically important traits in the breeding goal. The main objective of this study was to assess the relative performance of the traditional lactation average model (LAM) against the random regression test-day model (RRM) in the estimation of genetic parameters and prediction of breeding values for Holstein Friesian herds in Ethiopia. The data used consisted of 6,500 test-day (TD) records from 800 first-lactation Holstein Friesian cows that calved between 1997 and 2013. Co-variance components were estimated using the average information restricted maximum likelihood method under single trait animal model. The estimate of heritability for first-lactation milk yield was 0.30 from LAM whilst estimates from the RRM model ranged from 0.17 to 0.29 for the different stages of lactation. Genetic correlations between different TDs in first-lactation Holstein Friesian ranged from 0.37 to 0.99. The observed genetic correlation was less than unity between milk yields at different TDs, which indicated that the assumption of LAM may not be optimal for accurate evaluation of the genetic merit of animals. A close look at estimated breeding values from both models showed that RRM had higher standard deviation compared to LAM indicating that the TD model makes efficient utilization of TD information. Correlations of breeding values between models ranged from 0.90 to 0.96 for different group of sires and cows and marked re-rankings were observed in top sires and cows in moving from the traditional LAM to RRM evaluations. PMID:26194217

Calorie intake and patient outcomes in severe acute kidney injury: findings from The Randomized Evaluation of Normal vs. Augmented Level of Replacement Therapy (RENAL) study trial

PubMed Central

2014-01-01

Introduction Current practice in the delivery of caloric intake (DCI) in patients with severe acute kidney injury (AKI) receiving renal replacement therapy (RRT) is unknown. We aimed to describe calorie administration in patients enrolled in the Randomized Evaluation of Normal vs. Augmented Level of Replacement Therapy (RENAL) study and to assess the association between DCI and clinical outcomes. Methods We performed a secondary analysis in 1456 patients from the RENAL trial. We measured the dose and evolution of DCI during treatment and analyzed its association with major clinical outcomes using multivariable logistic regression, Cox proportional hazards models, and time adjusted models. Results Overall, mean DCI during treatment in ICU was low at only 10.9 ± 9 Kcal/kg/day for non-survivors and 11 ± 9 Kcal/kg/day for survivors. Among patients with a lower DCI (below the median) 334 of 729 (45.8%) had died at 90-days after randomization compared with 316 of 727 (43.3%) patients with a higher DCI (above the median) (P = 0.34). On multivariable logistic regression analysis, mean DCI carried an odds ratio of 0.95 (95% confidence interval (CI): 0.91-1.00; P = 0.06) per 100 Kcal increase for 90-day mortality. DCI was not associated with significant differences in renal replacement (RRT) free days, mechanical ventilation free days, ICU free days and hospital free days. These findings remained essentially unaltered after time adjusted analysis and Cox proportional hazards modeling. Conclusions In the RENAL study, mean DCI was low. Within the limits of such low caloric intake, greater DCI was not associated with improved clinical outcomes. Trial registration ClinicalTrials.gov number, NCT00221013 PMID:24629036

Joint genome-wide prediction in several populations accounting for randomness of genotypes: A hierarchical Bayes approach. I: Multivariate Gaussian priors for marker effects and derivation of the joint probability mass function of genotypes.

PubMed

Martínez, Carlos Alberto; Khare, Kshitij; Banerjee, Arunava; Elzo, Mauricio A

2017-03-21

It is important to consider heterogeneity of marker effects and allelic frequencies in across population genome-wide prediction studies. Moreover, all regression models used in genome-wide prediction overlook randomness of genotypes. In this study, a family of hierarchical Bayesian models to perform across population genome-wide prediction modeling genotypes as random variables and allowing population-specific effects for each marker was developed. Models shared a common structure and differed in the priors used and the assumption about residual variances (homogeneous or heterogeneous). Randomness of genotypes was accounted for by deriving the joint probability mass function of marker genotypes conditional on allelic frequencies and pedigree information. As a consequence, these models incorporated kinship and genotypic information that not only permitted to account for heterogeneity of allelic frequencies, but also to include individuals with missing genotypes at some or all loci without the need for previous imputation. This was possible because the non-observed fraction of the design matrix was treated as an unknown model parameter. For each model, a simpler version ignoring population structure, but still accounting for randomness of genotypes was proposed. Implementation of these models and computation of some criteria for model comparison were illustrated using two simulated datasets. Theoretical and computational issues along with possible applications, extensions and refinements were discussed. Some features of the models developed in this study make them promising for genome-wide prediction, the use of information contained in the probability distribution of genotypes is perhaps the most appealing. Further studies to assess the performance of the models proposed here and also to compare them with conventional models used in genome-wide prediction are needed. Copyright © 2017 Elsevier Ltd. All rights reserved.

Box–Cox Transformation and Random Regression Models for Fecal egg Count Data

PubMed Central

da Silva, Marcos Vinícius Gualberto Barbosa; Van Tassell, Curtis P.; Sonstegard, Tad S.; Cobuci, Jaime Araujo; Gasbarre, Louis C.

2012-01-01

Accurate genetic evaluation of livestock is based on appropriate modeling of phenotypic measurements. In ruminants, fecal egg count (FEC) is commonly used to measure resistance to nematodes. FEC values are not normally distributed and logarithmic transformations have been used in an effort to achieve normality before analysis. However, the transformed data are often still not normally distributed, especially when data are extremely skewed. A series of repeated FEC measurements may provide information about the population dynamics of a group or individual. A total of 6375 FEC measures were obtained for 410 animals between 1992 and 2003 from the Beltsville Agricultural Research Center Angus herd. Original data were transformed using an extension of the Box–Cox transformation to approach normality and to estimate (co)variance components. We also proposed using random regression models (RRM) for genetic and non-genetic studies of FEC. Phenotypes were analyzed using RRM and restricted maximum likelihood. Within the different orders of Legendre polynomials used, those with more parameters (order 4) adjusted FEC data best. Results indicated that the transformation of FEC data utilizing the Box–Cox transformation family was effective in reducing the skewness and kurtosis, and dramatically increased estimates of heritability, and measurements of FEC obtained in the period between 12 and 26 weeks in a 26-week experimental challenge period are genetically correlated. PMID:22303406

Box-Cox Transformation and Random Regression Models for Fecal egg Count Data.

PubMed

da Silva, Marcos Vinícius Gualberto Barbosa; Van Tassell, Curtis P; Sonstegard, Tad S; Cobuci, Jaime Araujo; Gasbarre, Louis C

2011-01-01

Accurate genetic evaluation of livestock is based on appropriate modeling of phenotypic measurements. In ruminants, fecal egg count (FEC) is commonly used to measure resistance to nematodes. FEC values are not normally distributed and logarithmic transformations have been used in an effort to achieve normality before analysis. However, the transformed data are often still not normally distributed, especially when data are extremely skewed. A series of repeated FEC measurements may provide information about the population dynamics of a group or individual. A total of 6375 FEC measures were obtained for 410 animals between 1992 and 2003 from the Beltsville Agricultural Research Center Angus herd. Original data were transformed using an extension of the Box-Cox transformation to approach normality and to estimate (co)variance components. We also proposed using random regression models (RRM) for genetic and non-genetic studies of FEC. Phenotypes were analyzed using RRM and restricted maximum likelihood. Within the different orders of Legendre polynomials used, those with more parameters (order 4) adjusted FEC data best. Results indicated that the transformation of FEC data utilizing the Box-Cox transformation family was effective in reducing the skewness and kurtosis, and dramatically increased estimates of heritability, and measurements of FEC obtained in the period between 12 and 26 weeks in a 26-week experimental challenge period are genetically correlated.

Using fixed-parameter and random-parameter ordered regression models to identify significant factors that affect the severity of drivers' injuries in vehicle-train collisions.

PubMed

Dabbour, Essam; Easa, Said; Haider, Murtaza

2017-10-01

This study attempts to identify significant factors that affect the severity of drivers' injuries when colliding with trains at railroad-grade crossings by analyzing the individual-specific heterogeneity related to those factors over a period of 15 years. Both fixed-parameter and random-parameter ordered regression models were used to analyze records of all vehicle-train collisions that occurred in the United States from January 1, 2001 to December 31, 2015. For fixed-parameter ordered models, both probit and negative log-log link functions were used. The latter function accounts for the fact that lower injury severity levels are more probable than higher ones. Separate models were developed for heavy and light-duty vehicles. Higher train and vehicle speeds, female, and young drivers (below the age of 21 years) were found to be consistently associated with higher severity of drivers' injuries for both heavy and light-duty vehicles. Furthermore, favorable weather, light-duty trucks (including pickup trucks, panel trucks, mini-vans, vans, and sports-utility vehicles), and senior drivers (above the age of 65 years) were found be consistently associated with higher severity of drivers' injuries for light-duty vehicles only. All other factors (e.g. air temperature, the type of warning devices, darkness conditions, and highway pavement type) were found to be temporally unstable, which may explain the conflicting findings of previous studies related to those factors. Copyright © 2017 Elsevier Ltd. All rights reserved.

How to derive biological information from the value of the normalization constant in allometric equations.

PubMed

Kaitaniemi, Pekka

2008-04-09

Allometric equations are widely used in many branches of biological science. The potential information content of the normalization constant b in allometric equations of the form Y = bX(a) has, however, remained largely neglected. To demonstrate the potential for utilizing this information, I generated a large number of artificial datasets that resembled those that are frequently encountered in biological studies, i.e., relatively small samples including measurement error or uncontrolled variation. The value of X was allowed to vary randomly within the limits describing different data ranges, and a was set to a fixed theoretical value. The constant b was set to a range of values describing the effect of a continuous environmental variable. In addition, a normally distributed random error was added to the values of both X and Y. Two different approaches were then used to model the data. The traditional approach estimated both a and b using a regression model, whereas an alternative approach set the exponent a at its theoretical value and only estimated the value of b. Both approaches produced virtually the same model fit with less than 0.3% difference in the coefficient of determination. Only the alternative approach was able to precisely reproduce the effect of the environmental variable, which was largely lost among noise variation when using the traditional approach. The results show how the value of b can be used as a source of valuable biological information if an appropriate regression model is selected.

The extension of total gain (TG) statistic in survival models: properties and applications.

PubMed

Choodari-Oskooei, Babak; Royston, Patrick; Parmar, Mahesh K B

2015-07-01

The results of multivariable regression models are usually summarized in the form of parameter estimates for the covariates, goodness-of-fit statistics, and the relevant p-values. These statistics do not inform us about whether covariate information will lead to any substantial improvement in prediction. Predictive ability measures can be used for this purpose since they provide important information about the practical significance of prognostic factors. R (2)-type indices are the most familiar forms of such measures in survival models, but they all have limitations and none is widely used. In this paper, we extend the total gain (TG) measure, proposed for a logistic regression model, to survival models and explore its properties using simulations and real data. TG is based on the binary regression quantile plot, otherwise known as the predictiveness curve. Standardised TG ranges from 0 (no explanatory power) to 1 ('perfect' explanatory power). The results of our simulations show that unlike many of the other R (2)-type predictive ability measures, TG is independent of random censoring. It increases as the effect of a covariate increases and can be applied to different types of survival models, including models with time-dependent covariate effects. We also apply TG to quantify the predictive ability of multivariable prognostic models developed in several disease areas. Overall, TG performs well in our simulation studies and can be recommended as a measure to quantify the predictive ability in survival models.

What's in a title? An assessment of whether randomized controlled trial in a title means that it is one.

PubMed

Koletsi, Despina; Pandis, Nikolaos; Polychronopoulou, Argy; Eliades, Theodore

2012-06-01

In this study, we aimed to investigate whether studies published in orthodontic journals and titled as randomized clinical trials are truly randomized clinical trials. A second objective was to explore the association of journal type and other publication characteristics on correct classification. American Journal of Orthodontics and Dentofacial Orthopedics, European Journal of Orthodontics, Angle Orthodontist, Journal of Orthodontics, Orthodontics and Craniofacial Research, World Journal of Orthodontics, Australian Orthodontic Journal, and Journal of Orofacial Orthopedics were hand searched for clinical trials labeled in the title as randomized from 1979 to July 2011. The data were analyzed by using descriptive statistics, and univariable and multivariable examinations of statistical associations via ordinal logistic regression modeling (proportional odds model). One hundred twelve trials were identified. Of the included trials, 33 (29.5%) were randomized clinical trials, 52 (46.4%) had an unclear status, and 27 (24.1%) were not randomized clinical trials. In the multivariable analysis among the included journal types, year of publication, number of authors, multicenter trial, and involvement of statistician were significant predictors of correctly classifying a study as a randomized clinical trial vs unclear and not a randomized clinical trial. From 112 clinical trials in the orthodontic literature labeled as randomized clinical trials, only 29.5% were identified as randomized clinical trials based on clear descriptions of appropriate random number generation and allocation concealment. The type of journal, involvement of a statistician, multicenter trials, greater numbers of authors, and publication year were associated with correct clinical trial classification. This study indicates the need of clear and accurate reporting of clinical trials and the need for educating investigators on randomized clinical trial methodology. Copyright © 2012 American Association of Orthodontists. Published by Mosby, Inc. All rights reserved.

Evaluating the Effectiveness of Developmental Mathematics by Embedding a Randomized Experiment within a Regression Discontinuity Design

ERIC Educational Resources Information Center

Moss, Brian G.; Yeaton, William H.; Lloyd, Jane E.

2014-01-01

Using a novel design approach, a randomized experiment (RE) was embedded within a regression discontinuity (RD) design (R-RE-D) to evaluate the impact of developmental mathematics at a large midwestern college ("n" = 2,122). Within a region of uncertainty near the cut-score, estimates of benefit from a prospective RE were closely…

Fitting parametric random effects models in very large data sets with application to VHA national data

PubMed Central

2012-01-01

Background With the current focus on personalized medicine, patient/subject level inference is often of key interest in translational research. As a result, random effects models (REM) are becoming popular for patient level inference. However, for very large data sets that are characterized by large sample size, it can be difficult to fit REM using commonly available statistical software such as SAS since they require inordinate amounts of computer time and memory allocations beyond what are available preventing model convergence. For example, in a retrospective cohort study of over 800,000 Veterans with type 2 diabetes with longitudinal data over 5 years, fitting REM via generalized linear mixed modeling using currently available standard procedures in SAS (e.g. PROC GLIMMIX) was very difficult and same problems exist in Stata’s gllamm or R’s lme packages. Thus, this study proposes and assesses the performance of a meta regression approach and makes comparison with methods based on sampling of the full data. Data We use both simulated and real data from a national cohort of Veterans with type 2 diabetes (n=890,394) which was created by linking multiple patient and administrative files resulting in a cohort with longitudinal data collected over 5 years. Methods and results The outcome of interest was mean annual HbA1c measured over a 5 years period. Using this outcome, we compared parameter estimates from the proposed random effects meta regression (REMR) with estimates based on simple random sampling and VISN (Veterans Integrated Service Networks) based stratified sampling of the full data. Our results indicate that REMR provides parameter estimates that are less likely to be biased with tighter confidence intervals when the VISN level estimates are homogenous. Conclusion When the interest is to fit REM in repeated measures data with very large sample size, REMR can be used as a good alternative. It leads to reasonable inference for both Gaussian and non-Gaussian responses if parameter estimates are homogeneous across VISNs. PMID:23095325

Mapping of the DLQI scores to EQ-5D utility values using ordinal logistic regression.

PubMed

Ali, Faraz Mahmood; Kay, Richard; Finlay, Andrew Y; Piguet, Vincent; Kupfer, Joerg; Dalgard, Florence; Salek, M Sam

2017-11-01

The Dermatology Life Quality Index (DLQI) and the European Quality of Life-5 Dimension (EQ-5D) are separate measures that may be used to gather health-related quality of life (HRQoL) information from patients. The EQ-5D is a generic measure from which health utility estimates can be derived, whereas the DLQI is a specialty-specific measure to assess HRQoL. To reduce the burden of multiple measures being administered and to enable a more disease-specific calculation of health utility estimates, we explored an established mathematical technique known as ordinal logistic regression (OLR) to develop an appropriate model to map DLQI data to EQ-5D-based health utility estimates. Retrospective data from 4010 patients were randomly divided five times into two groups for the derivation and testing of the mapping model. Split-half cross-validation was utilized resulting in a total of ten ordinal logistic regression models for each of the five EQ-5D dimensions against age, sex, and all ten items of the DLQI. Using Monte Carlo simulation, predicted health utility estimates were derived and compared against those observed. This method was repeated for both OLR and a previously tested mapping methodology based on linear regression. The model was shown to be highly predictive and its repeated fitting demonstrated a stable model using OLR as well as linear regression. The mean differences between OLR-predicted health utility estimates and observed health utility estimates ranged from 0.0024 to 0.0239 across the ten modeling exercises, with an average overall difference of 0.0120 (a 1.6% underestimate, not of clinical importance). This modeling framework developed in this study will enable researchers to calculate EQ-5D health utility estimates from a specialty-specific study population, reducing patient and economic burden.

Connections between survey calibration estimators and semiparametric models for incomplete data

PubMed Central

Lumley, Thomas; Shaw, Pamela A.; Dai, James Y.

2012-01-01

Survey calibration (or generalized raking) estimators are a standard approach to the use of auxiliary information in survey sampling, improving on the simple Horvitz–Thompson estimator. In this paper we relate the survey calibration estimators to the semiparametric incomplete-data estimators of Robins and coworkers, and to adjustment for baseline variables in a randomized trial. The development based on calibration estimators explains the ‘estimated weights’ paradox and provides useful heuristics for constructing practical estimators. We present some examples of using calibration to gain precision without making additional modelling assumptions in a variety of regression models. PMID:23833390

Statistically extracted fundamental watershed variables for estimating the loads of total nitrogen in small streams

USGS Publications Warehouse

Kronholm, Scott C.; Capel, Paul D.; Terziotti, Silvia

2016-01-01

Accurate estimation of total nitrogen loads is essential for evaluating conditions in the aquatic environment. Extrapolation of estimates beyond measured streams will greatly expand our understanding of total nitrogen loading to streams. Recursive partitioning and random forest regression were used to assess 85 geospatial, environmental, and watershed variables across 636 small (<585 km2) watersheds to determine which variables are fundamentally important to the estimation of annual loads of total nitrogen. Initial analysis led to the splitting of watersheds into three groups based on predominant land use (agricultural, developed, and undeveloped). Nitrogen application, agricultural and developed land area, and impervious or developed land in the 100-m stream buffer were commonly extracted variables by both recursive partitioning and random forest regression. A series of multiple linear regression equations utilizing the extracted variables were created and applied to the watersheds. As few as three variables explained as much as 76 % of the variability in total nitrogen loads for watersheds with predominantly agricultural land use. Catchment-scale national maps were generated to visualize the total nitrogen loads and yields across the USA. The estimates provided by these models can inform water managers and help identify areas where more in-depth monitoring may be beneficial.

Cross-Sectional HIV Incidence Surveillance: A Benchmarking of Approaches for Estimating the 'Mean Duration of Recent Infection'.

PubMed

Kassanjee, Reshma; De Angelis, Daniela; Farah, Marian; Hanson, Debra; Labuschagne, Jan Phillipus Lourens; Laeyendecker, Oliver; Le Vu, Stéphane; Tom, Brian; Wang, Rui; Welte, Alex

2017-03-01

The application of biomarkers for 'recent' infection in cross-sectional HIV incidence surveillance requires the estimation of critical biomarker characteristics. Various approaches have been employed for using longitudinal data to estimate the Mean Duration of Recent Infection (MDRI) - the average time in the 'recent' state. In this systematic benchmarking of MDRI estimation approaches, a simulation platform was used to measure accuracy and precision of over twenty approaches, in thirty scenarios capturing various study designs, subject behaviors and test dynamics that may be encountered in practice. Results highlight that assuming a single continuous sojourn in the 'recent' state can produce substantial bias. Simple interpolation provides useful MDRI estimates provided subjects are tested at regular intervals. Regression performs the best - while 'random effects' describe the subject-clustering in the data, regression models without random effects proved easy to implement, stable, and of similar accuracy in scenarios considered; robustness to parametric assumptions was improved by regressing 'recent'/'non-recent' classifications rather than continuous biomarker readings. All approaches were vulnerable to incorrect assumptions about subjects' (unobserved) infection times. Results provided show the relationships between MDRI estimation performance and the number of subjects, inter-visit intervals, missed visits, loss to follow-up, and aspects of biomarker signal and noise.

Analysis of Machine Learning Techniques for Heart Failure Readmissions.

PubMed

Mortazavi, Bobak J; Downing, Nicholas S; Bucholz, Emily M; Dharmarajan, Kumar; Manhapra, Ajay; Li, Shu-Xia; Negahban, Sahand N; Krumholz, Harlan M

2016-11-01

The current ability to predict readmissions in patients with heart failure is modest at best. It is unclear whether machine learning techniques that address higher dimensional, nonlinear relationships among variables would enhance prediction. We sought to compare the effectiveness of several machine learning algorithms for predicting readmissions. Using data from the Telemonitoring to Improve Heart Failure Outcomes trial, we compared the effectiveness of random forests, boosting, random forests combined hierarchically with support vector machines or logistic regression (LR), and Poisson regression against traditional LR to predict 30- and 180-day all-cause readmissions and readmissions because of heart failure. We randomly selected 50% of patients for a derivation set, and a validation set comprised the remaining patients, validated using 100 bootstrapped iterations. We compared C statistics for discrimination and distributions of observed outcomes in risk deciles for predictive range. In 30-day all-cause readmission prediction, the best performing machine learning model, random forests, provided a 17.8% improvement over LR (mean C statistics, 0.628 and 0.533, respectively). For readmissions because of heart failure, boosting improved the C statistic by 24.9% over LR (mean C statistic 0.678 and 0.543, respectively). For 30-day all-cause readmission, the observed readmission rates in the lowest and highest deciles of predicted risk with random forests (7.8% and 26.2%, respectively) showed a much wider separation than LR (14.2% and 16.4%, respectively). Machine learning methods improved the prediction of readmission after hospitalization for heart failure compared with LR and provided the greatest predictive range in observed readmission rates. © 2016 American Heart Association, Inc.

Sensitivity Analysis of Mechanical Parameters of Different Rock Layers to the Stability of Coal Roadway in Soft Rock Strata

PubMed Central

Zhao, Zeng-hui; Wang, Wei-ming; Gao, Xin; Yan, Ji-xing

2013-01-01

According to the geological characteristics of Xinjiang Ili mine in western area of China, a physical model of interstratified strata composed of soft rock and hard coal seam was established. Selecting the tunnel position, deformation modulus, and strength parameters of each layer as influencing factors, the sensitivity coefficient of roadway deformation to each parameter was firstly analyzed based on a Mohr-Columb strain softening model and nonlinear elastic-plastic finite element analysis. Then the effect laws of influencing factors which showed high sensitivity were further discussed. Finally, a regression model for the relationship between roadway displacements and multifactors was obtained by equivalent linear regression under multiple factors. The results show that the roadway deformation is highly sensitive to the depth of coal seam under the floor which should be considered in the layout of coal roadway; deformation modulus and strength of coal seam and floor have a great influence on the global stability of tunnel; on the contrary, roadway deformation is not sensitive to the mechanical parameters of soft roof; roadway deformation under random combinations of multi-factors can be deduced by the regression model. These conclusions provide theoretical significance to the arrangement and stability maintenance of coal roadway. PMID:24459447

Basic Diagnosis and Prediction of Persistent Contrail Occurrence using High-resolution Numerical Weather Analyses/Forecasts and Logistic Regression. Part I: Effects of Random Error

NASA Technical Reports Server (NTRS)

Duda, David P.; Minnis, Patrick

2009-01-01

Straightforward application of the Schmidt-Appleman contrail formation criteria to diagnose persistent contrail occurrence from numerical weather prediction data is hindered by significant bias errors in the upper tropospheric humidity. Logistic models of contrail occurrence have been proposed to overcome this problem, but basic questions remain about how random measurement error may affect their accuracy. A set of 5000 synthetic contrail observations is created to study the effects of random error in these probabilistic models. The simulated observations are based on distributions of temperature, humidity, and vertical velocity derived from Advanced Regional Prediction System (ARPS) weather analyses. The logistic models created from the simulated observations were evaluated using two common statistical measures of model accuracy, the percent correct (PC) and the Hanssen-Kuipers discriminant (HKD). To convert the probabilistic results of the logistic models into a dichotomous yes/no choice suitable for the statistical measures, two critical probability thresholds are considered. The HKD scores are higher when the climatological frequency of contrail occurrence is used as the critical threshold, while the PC scores are higher when the critical probability threshold is 0.5. For both thresholds, typical random errors in temperature, relative humidity, and vertical velocity are found to be small enough to allow for accurate logistic models of contrail occurrence. The accuracy of the models developed from synthetic data is over 85 percent for both the prediction of contrail occurrence and non-occurrence, although in practice, larger errors would be anticipated.

A multilevel model to estimate the within- and the between-center components of the exposure/disease association in the EPIC study.

PubMed

Sera, Francesco; Ferrari, Pietro

2015-01-01

In a multicenter study, the overall relationship between exposure and the risk of cancer can be broken down into a within-center component, which reflects the individual level association, and a between-center relationship, which captures the association at the aggregate level. A piecewise exponential proportional hazards model with random effects was used to evaluate the association between dietary fiber intake and colorectal cancer (CRC) risk in the EPIC study. During an average follow-up of 11.0 years, 4,517 CRC events occurred among study participants recruited in 28 centers from ten European countries. Models were adjusted by relevant confounding factors. Heterogeneity among centers was modelled with random effects. Linear regression calibration was used to account for errors in dietary questionnaire (DQ) measurements. Risk ratio estimates for a 10 g/day increment in dietary fiber were equal to 0.90 (95%CI: 0.85, 0.96) and 0.85 (0.64, 1.14), at the individual and aggregate levels, respectively, while calibrated estimates were 0.85 (0.76, 0.94), and 0.87 (0.65, 1.15), respectively. In multicenter studies, over a straightforward ecological analysis, random effects models allow information at the individual and ecologic levels to be captured, while controlling for confounding at both levels of evidence.

Predicting active-layer soil thickness using topographic variables at a small watershed scale

PubMed Central

Li, Aidi; Tan, Xing; Wu, Wei; Liu, Hongbin; Zhu, Jie

2017-01-01

Knowledge about the spatial distribution of active-layer (AL) soil thickness is indispensable for ecological modeling, precision agriculture, and land resource management. However, it is difficult to obtain the details on AL soil thickness by using conventional soil survey method. In this research, the objective is to investigate the possibility and accuracy of mapping the spatial distribution of AL soil thickness through random forest (RF) model by using terrain variables at a small watershed scale. A total of 1113 soil samples collected from the slope fields were randomly divided into calibration (770 soil samples) and validation (343 soil samples) sets. Seven terrain variables including elevation, aspect, relative slope position, valley depth, flow path length, slope height, and topographic wetness index were derived from a digital elevation map (30 m). The RF model was compared with multiple linear regression (MLR), geographically weighted regression (GWR) and support vector machines (SVM) approaches based on the validation set. Model performance was evaluated by precision criteria of mean error (ME), mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R2). Comparative results showed that RF outperformed MLR, GWR and SVM models. The RF gave better values of ME (0.39 cm), MAE (7.09 cm), and RMSE (10.85 cm) and higher R2 (62%). The sensitivity analysis demonstrated that the DEM had less uncertainty than the AL soil thickness. The outcome of the RF model indicated that elevation, flow path length and valley depth were the most important factors affecting the AL soil thickness variability across the watershed. These results demonstrated the RF model is a promising method for predicting spatial distribution of AL soil thickness using terrain parameters. PMID:28877196

Soil salinity study in Northern Great Plains sodium affected soil

NASA Astrophysics Data System (ADS)

Kharel, Tulsi P.

Climate and land-use changes when combined with the marine sediments that underlay portions of the Northern Great Plains have increased the salinization and sodification risks. The objectives of this dissertation were to compare three chemical amendments (calcium chloride, sulfuric acid and gypsum) remediation strategies on water permeability and sodium (Na) transport in undisturbed soil columns and to develop a remote sensing technique to characterize salinization in South Dakota soils. Forty-eight undisturbed soil columns (30 cm x 15 cm) collected from White Lake, Redfield, and Pierpont were used to assess the chemical remediation strategies. In this study the experimental design was a completely randomized design and each treatment was replicated four times. Following the application of chemical remediation strategies, 45.2 cm of water was leached through these columns. The leachate was separated into 120- ml increments and analyzed for Na and electrical conductivity (EC). Sulfuric acid increased Na leaching, whereas gypsum and CaCl2 increased water permeability. Our results further indicate that to maintain effective water permeability, ratio between soil EC and sodium absorption ratio (SAR) should be considered. In the second study, soil samples from 0-15 cm depth in 62 x 62 m grid spacing were taken from the South Dakota Pierpont (65 ha) and Redfield (17 ha) sites. Saturated paste EC was measured on each soil sample. At each sampling points reflectance and derived indices (Landsat 5, 7, 8 images), elevation, slope and aspect (LiDAR) were extracted. Regression models based on multiple linear regression, classification and regression tree, cubist, and random forest techniques were developed and their ability to predict soil EC were compared. Results showed that: 1) Random forest method was found to be the most effective method because of its ability to capture spatially correlated variation, 2) the short wave infrared (1.5 -2.29 mum) and near infrared (0.75-0.90 mum) were very sensitive to soil salinity; 3) EC prediction model using all 3 season (spring, summer and fall) images was better on state wide validation dataset compared to individual season model. Finally, in eastern South Dakota, the model predicted that from 2008 to 2012, EC increased in 569,165 ha or 13.4% of the land seeded to corn (Zea mays L.) or soybeans (Glycine max L).

Nonconvex Sparse Logistic Regression With Weakly Convex Regularization

NASA Astrophysics Data System (ADS)

Shen, Xinyue; Gu, Yuantao

2018-06-01

In this work we propose to fit a sparse logistic regression model by a weakly convex regularized nonconvex optimization problem. The idea is based on the finding that a weakly convex function as an approximation of the $\\ell_0$ pseudo norm is able to better induce sparsity than the commonly used $\\ell_1$ norm. For a class of weakly convex sparsity inducing functions, we prove the nonconvexity of the corresponding sparse logistic regression problem, and study its local optimality conditions and the choice of the regularization parameter to exclude trivial solutions. Despite the nonconvexity, a method based on proximal gradient descent is used to solve the general weakly convex sparse logistic regression, and its convergence behavior is studied theoretically. Then the general framework is applied to a specific weakly convex function, and a necessary and sufficient local optimality condition is provided. The solution method is instantiated in this case as an iterative firm-shrinkage algorithm, and its effectiveness is demonstrated in numerical experiments by both randomly generated and real datasets.

A comparison of adaptive sampling designs and binary spatial models: A simulation study using a census of Bromus inermis

USGS Publications Warehouse

Irvine, Kathryn M.; Thornton, Jamie; Backus, Vickie M.; Hohmann, Matthew G.; Lehnhoff, Erik A.; Maxwell, Bruce D.; Michels, Kurt; Rew, Lisa

2013-01-01

Commonly in environmental and ecological studies, species distribution data are recorded as presence or absence throughout a spatial domain of interest. Field based studies typically collect observations by sampling a subset of the spatial domain. We consider the effects of six different adaptive and two non-adaptive sampling designs and choice of three binary models on both predictions to unsampled locations and parameter estimation of the regression coefficients (species–environment relationships). Our simulation study is unique compared to others to date in that we virtually sample a true known spatial distribution of a nonindigenous plant species, Bromus inermis. The census of B. inermis provides a good example of a species distribution that is both sparsely (1.9 % prevalence) and patchily distributed. We find that modeling the spatial correlation using a random effect with an intrinsic Gaussian conditionally autoregressive prior distribution was equivalent or superior to Bayesian autologistic regression in terms of predicting to un-sampled areas when strip adaptive cluster sampling was used to survey B. inermis. However, inferences about the relationships between B. inermis presence and environmental predictors differed between the two spatial binary models. The strip adaptive cluster designs we investigate provided a significant advantage in terms of Markov chain Monte Carlo chain convergence when trying to model a sparsely distributed species across a large area. In general, there was little difference in the choice of neighborhood, although the adaptive king was preferred when transects were randomly placed throughout the spatial domain.

Policy Implications and Suggestions on Administrative Measures of Urban Flood

NASA Astrophysics Data System (ADS)

Lee, S. V.; Lee, M. J.; Lee, C.; Yoon, J. H.; Chae, S. H.

2017-12-01

The frequency and intensity of floods are increasing worldwide as recent climate change progresses gradually. Flood management should be policy-oriented in urban municipalities due to the characteristics of urban areas with a lot of damage. Therefore, the purpose of this study is to prepare a flood susceptibility map by using data mining model and make a policy suggestion on administrative measures of urban flood. Therefore, we constructed a spatial database by collecting relevant factors including the topography, geology, soil and land use data of the representative city, Seoul, the capital city of Korea. Flood susceptibility map was constructed by applying the data mining models of random forest and boosted tree model to input data and existing flooded area data in 2010. The susceptibility map has been validated using the 2011 flood area data which was not used for training. The predictor importance value of each factor to the results was calculated in this process. The distance from the water, DEM and geology showed a high predictor importance value which means to be a high priority for flood preparation policy. As a result of receiver operating characteristic (ROC), random forest model showed 78.78% and 79.18% accuracy of regression and classification and boosted tree model showed 77.55% and 77.26% accuracy of regression and classification, respectively. The results show that the flood susceptibility maps can be applied to flood prevention and management, and it also can help determine the priority areas for flood mitigation policy by providing useful information to policy makers.

Predicting diabetes mellitus using SMOTE and ensemble machine learning approach: The Henry Ford ExercIse Testing (FIT) project.

PubMed

Alghamdi, Manal; Al-Mallah, Mouaz; Keteyian, Steven; Brawner, Clinton; Ehrman, Jonathan; Sakr, Sherif

2017-01-01

Machine learning is becoming a popular and important approach in the field of medical research. In this study, we investigate the relative performance of various machine learning methods such as Decision Tree, Naïve Bayes, Logistic Regression, Logistic Model Tree and Random Forests for predicting incident diabetes using medical records of cardiorespiratory fitness. In addition, we apply different techniques to uncover potential predictors of diabetes. This FIT project study used data of 32,555 patients who are free of any known coronary artery disease or heart failure who underwent clinician-referred exercise treadmill stress testing at Henry Ford Health Systems between 1991 and 2009 and had a complete 5-year follow-up. At the completion of the fifth year, 5,099 of those patients have developed diabetes. The dataset contained 62 attributes classified into four categories: demographic characteristics, disease history, medication use history, and stress test vital signs. We developed an Ensembling-based predictive model using 13 attributes that were selected based on their clinical importance, Multiple Linear Regression, and Information Gain Ranking methods. The negative effect of the imbalance class of the constructed model was handled by Synthetic Minority Oversampling Technique (SMOTE). The overall performance of the predictive model classifier was improved by the Ensemble machine learning approach using the Vote method with three Decision Trees (Naïve Bayes Tree, Random Forest, and Logistic Model Tree) and achieved high accuracy of prediction (AUC = 0.92). The study shows the potential of ensembling and SMOTE approaches for predicting incident diabetes using cardiorespiratory fitness data.

Predicting Coastal Flood Severity using Random Forest Algorithm

NASA Astrophysics Data System (ADS)

Sadler, J. M.; Goodall, J. L.; Morsy, M. M.; Spencer, K.

2017-12-01

Coastal floods have become more common recently and are predicted to further increase in frequency and severity due to sea level rise. Predicting floods in coastal cities can be difficult due to the number of environmental and geographic factors which can influence flooding events. Built stormwater infrastructure and irregular urban landscapes add further complexity. This paper demonstrates the use of machine learning algorithms in predicting street flood occurrence in an urban coastal setting. The model is trained and evaluated using data from Norfolk, Virginia USA from September 2010 - October 2016. Rainfall, tide levels, water table levels, and wind conditions are used as input variables. Street flooding reports made by city workers after named and unnamed storm events, ranging from 1-159 reports per event, are the model output. Results show that Random Forest provides predictive power in estimating the number of flood occurrences given a set of environmental conditions with an out-of-bag root mean squared error of 4.3 flood reports and a mean absolute error of 0.82 flood reports. The Random Forest algorithm performed much better than Poisson regression. From the Random Forest model, total daily rainfall was by far the most important factor in flood occurrence prediction, followed by daily low tide and daily higher high tide. The model demonstrated here could be used to predict flood severity based on forecast rainfall and tide conditions and could be further enhanced using more complete street flooding data for model training.

Incorporating local land use regression and satellite aerosol optical depth in a hybrid model of spatiotemporal PM2.5 exposures in the Mid-Atlantic states.

PubMed

Kloog, Itai; Nordio, Francesco; Coull, Brent A; Schwartz, Joel

2012-11-06

Satellite-derived aerosol optical depth (AOD) measurements have the potential to provide spatiotemporally resolved predictions of both long and short-term exposures, but previous studies have generally shown moderate predictive power and lacked detailed high spatio- temporal resolution predictions across large domains. We aimed at extending our previous work by validating our model in another region with different geographical and metrological characteristics, and incorporating fine scale land use regression and nonrandom missingness to better predict PM(2.5) concentrations for days with or without satellite AOD measures. We start by calibrating AOD data for 2000-2008 across the Mid-Atlantic. We used mixed models regressing PM(2.5) measurements against day-specific random intercepts, and fixed and random AOD and temperature slopes. We used inverse probability weighting to account for nonrandom missingness of AOD, nested regions within days to capture spatial variation in the daily calibration, and introduced a penalization method that reduces the dimensionality of the large number of spatial and temporal predictors without selecting different predictors in different locations. We then take advantage of the association between grid-cell specific AOD values and PM(2.5) monitoring data, together with associations between AOD values in neighboring grid cells to develop grid cell predictions when AOD is missing. Finally to get local predictions (at the resolution of 50 m), we regressed the residuals from the predictions for each monitor from these previous steps against the local land use variables specific for each monitor. "Out-of-sample" 10-fold cross-validation was used to quantify the accuracy of our predictions at each step. For all days without AOD values, model performance was excellent (mean "out-of-sample" R(2) = 0.81, year-to-year variation 0.79-0.84). Upon removal of outliers in the PM(2.5) monitoring data, the results of the cross validation procedure was even better (overall mean "out of sample"R(2) of 0.85). Further, cross validation results revealed no bias in the predicted concentrations (Slope of observed vs predicted = 0.97-1.01). Our model allows one to reliably assess short-term and long-term human exposures in order to investigate both the acute and effects of ambient particles, respectively.

A comparison of model-based imputation methods for handling missing predictor values in a linear regression model: A simulation study

NASA Astrophysics Data System (ADS)

Hasan, Haliza; Ahmad, Sanizah; Osman, Balkish Mohd; Sapri, Shamsiah; Othman, Nadirah

2017-08-01

In regression analysis, missing covariate data has been a common problem. Many researchers use ad hoc methods to overcome this problem due to the ease of implementation. However, these methods require assumptions about the data that rarely hold in practice. Model-based methods such as Maximum Likelihood (ML) using the expectation maximization (EM) algorithm and Multiple Imputation (MI) are more promising when dealing with difficulties caused by missing data. Then again, inappropriate methods of missing value imputation can lead to serious bias that severely affects the parameter estimates. The main objective of this study is to provide a better understanding regarding missing data concept that can assist the researcher to select the appropriate missing data imputation methods. A simulation study was performed to assess the effects of different missing data techniques on the performance of a regression model. The covariate data were generated using an underlying multivariate normal distribution and the dependent variable was generated as a combination of explanatory variables. Missing values in covariate were simulated using a mechanism called missing at random (MAR). Four levels of missingness (10%, 20%, 30% and 40%) were imposed. ML and MI techniques available within SAS software were investigated. A linear regression analysis was fitted and the model performance measures; MSE, and R-Squared were obtained. Results of the analysis showed that MI is superior in handling missing data with highest R-Squared and lowest MSE when percent of missingness is less than 30%. Both methods are unable to handle larger than 30% level of missingness.

A novel prediction approach for antimalarial activities of Trimethoprim, Pyrimethamine, and Cycloguanil analogues using extremely randomized trees.

PubMed

Nattee, Cholwich; Khamsemanan, Nirattaya; Lawtrakul, Luckhana; Toochinda, Pisanu; Hannongbua, Supa

2017-01-01

Malaria is still one of the most serious diseases in tropical regions. This is due in part to the high resistance against available drugs for the inhibition of parasites, Plasmodium, the cause of the disease. New potent compounds with high clinical utility are urgently needed. In this work, we created a novel model using a regression tree to study structure-activity relationships and predict the inhibition constant, K i of three different antimalarial analogues (Trimethoprim, Pyrimethamine, and Cycloguanil) based on their molecular descriptors. To the best of our knowledge, this work is the first attempt to study the structure-activity relationships of all three analogues combined. The most relevant descriptors and appropriate parameters of the regression tree are harvested using extremely randomized trees. These descriptors are water accessible surface area, Log of the aqueous solubility, total hydrophobic van der Waals surface area, and molecular refractivity. Out of all possible combinations of these selected parameters and descriptors, the tree with the strongest coefficient of determination is selected to be our prediction model. Predicted K i values from the proposed model show a strong coefficient of determination, R 2 =0.996, to experimental K i values. From the structure of the regression tree, compounds with high accessible surface area of all hydrophobic atoms (ASA_H) and low aqueous solubility of inhibitors (Log S) generally possess low K i values. Our prediction model can also be utilized as a screening test for new antimalarial drug compounds which may reduce the time and expenses for new drug development. New compounds with high predicted K i should be excluded from further drug development. It is also our inference that a threshold of ASA_H greater than 575.80 and Log S less than or equal to -4.36 is a sufficient condition for a new compound to possess a low K i . Copyright © 2016 Elsevier Inc. All rights reserved.

Systematic review using meta-analyses to estimate dose-response relationships between iodine intake and biomarkers of iodine status in different population groups.

PubMed

Ristić-Medić, Danijela; Dullemeijer, Carla; Tepsić, Jasna; Petrović-Oggiano, Gordana; Popović, Tamara; Arsić, Aleksandra; Glibetić, Marija; Souverein, Olga W; Collings, Rachel; Cavelaars, Adriënne; de Groot, Lisette; van't Veer, Pieter; Gurinović, Mirjana

2014-03-01

The objective of this systematic review was to identify studies investigating iodine intake and biomarkers of iodine status, to assess the data of the selected studies, and to estimate dose-response relationships using meta-analysis. All randomized controlled trials, prospective cohort studies, nested case-control studies, and cross-sectional studies that supplied or measured dietary iodine and measured iodine biomarkers were included. The overall pooled regression coefficient (β) and the standard error of β were calculated by random-effects meta-analysis on a double-log scale, using the calculated intake-status regression coefficient (β) for each individual study. The results of pooled randomized controlled trials indicated that the doubling of dietary iodine intake increased urinary iodine concentrations by 14% in children and adolescents, by 57% in adults and the elderly, and by 81% in pregnant women. The dose-response relationship between iodine intake and biomarkers of iodine status indicated a 12% decrease in thyroid-stimulating hormone and a 31% decrease in thyroglobulin in pregnant women. The model of dose-response quantification used to describe the relationship between iodine intake and biomarkers of iodine status may be useful for providing complementary evidence to support recommendations for iodine intake in different population groups.

Metabolic Profiling of Adiponectin Levels in Adults: Mendelian Randomization Analysis.

PubMed

Borges, Maria Carolina; Barros, Aluísio J D; Ferreira, Diana L Santos; Casas, Juan Pablo; Horta, Bernardo Lessa; Kivimaki, Mika; Kumari, Meena; Menon, Usha; Gaunt, Tom R; Ben-Shlomo, Yoav; Freitas, Deise F; Oliveira, Isabel O; Gentry-Maharaj, Aleksandra; Fourkala, Evangelia; Lawlor, Debbie A; Hingorani, Aroon D

2017-12-01

Adiponectin, a circulating adipocyte-derived protein, has insulin-sensitizing, anti-inflammatory, antiatherogenic, and cardiomyocyte-protective properties in animal models. However, the systemic effects of adiponectin in humans are unknown. Our aims were to define the metabolic profile associated with higher blood adiponectin concentration and investigate whether variation in adiponectin concentration affects the systemic metabolic profile. We applied multivariable regression in ≤5909 adults and Mendelian randomization (using cis -acting genetic variants in the vicinity of the adiponectin gene as instrumental variables) for analyzing the causal effect of adiponectin in the metabolic profile of ≤37 545 adults. Participants were largely European from 6 longitudinal studies and 1 genome-wide association consortium. In the multivariable regression analyses, higher circulating adiponectin was associated with higher high-density lipoprotein lipids and lower very-low-density lipoprotein lipids, glucose levels, branched-chain amino acids, and inflammatory markers. However, these findings were not supported by Mendelian randomization analyses for most metabolites. Findings were consistent between sexes and after excluding high-risk groups (defined by age and occurrence of previous cardiovascular event) and 1 study with admixed population. Our findings indicate that blood adiponectin concentration is more likely to be an epiphenomenon in the context of metabolic disease than a key determinant. © 2017 The Authors.

Random Measurement Error Does Not Bias the Treatment Effect Estimate in the Regression-Discontinuity Design. II. When an Interaction Effect Is Present.

ERIC Educational Resources Information Center

Trochim, William M. K.; And Others

1991-01-01

The regression-discontinuity design involving a treatment interaction effect (TIE), pretest-posttest functional form specification, and choice of point-of-estimation of the TIE are examined. Formulas for controlling the magnitude of TIE in simulations can be used for simulating the randomized experimental case where estimation is not at the…

VARIABLE SELECTION FOR REGRESSION MODELS WITH MISSING DATA

PubMed Central

Garcia, Ramon I.; Ibrahim, Joseph G.; Zhu, Hongtu

2009-01-01

We consider the variable selection problem for a class of statistical models with missing data, including missing covariate and/or response data. We investigate the smoothly clipped absolute deviation penalty (SCAD) and adaptive LASSO and propose a unified model selection and estimation procedure for use in the presence of missing data. We develop a computationally attractive algorithm for simultaneously optimizing the penalized likelihood function and estimating the penalty parameters. Particularly, we propose to use a model selection criterion, called the ICQ statistic, for selecting the penalty parameters. We show that the variable selection procedure based on ICQ automatically and consistently selects the important covariates and leads to efficient estimates with oracle properties. The methodology is very general and can be applied to numerous situations involving missing data, from covariates missing at random in arbitrary regression models to nonignorably missing longitudinal responses and/or covariates. Simulations are given to demonstrate the methodology and examine the finite sample performance of the variable selection procedures. Melanoma data from a cancer clinical trial is presented to illustrate the proposed methodology. PMID:20336190

Genetic analysis of fat-to-protein ratio, milk yield and somatic cell score of Holstein cows in Japan in the first three lactations by using a random regression model.

PubMed

Nishiura, Akiko; Sasaki, Osamu; Aihara, Mitsuo; Takeda, Hisato; Satoh, Masahiro

2015-12-01

We estimated the genetic parameters of fat-to-protein ratio (FPR) and the genetic correlations between FPR and milk yield or somatic cell score in the first three lactations in dairy cows. Data included 3,079,517 test-day records of 201,138 Holstein cows in Japan from 2006 to 2011. Genetic parameters were estimated with a multiple-trait random regression model in which the records within and between parities were treated as separate traits. The phenotypic values of FPR increased soon after parturition and peaked at 10 to 20 days in milk, then decreased slowly in mid- and late lactation. Heritability estimates for FPR yielded moderate values. Genetic correlations of FPR among parities were low in early lactation. Genetic correlations between FPR and milk yield were positive and low in early lactation, but only in the first lactation. Genetic correlations between FPR and somatic cell score were positive in early lactation and decreased to become negative in mid- to late lactation. By using these results for genetic evaluation it should be possible to improve energy balance in dairy cows. © 2015 Japanese Society of Animal Science.

Prediction of In-hospital Mortality in Emergency Department Patients With Sepsis: A Local Big Data-Driven, Machine Learning Approach.

PubMed

Taylor, R Andrew; Pare, Joseph R; Venkatesh, Arjun K; Mowafi, Hani; Melnick, Edward R; Fleischman, William; Hall, M Kennedy

2016-03-01

Predictive analytics in emergency care has mostly been limited to the use of clinical decision rules (CDRs) in the form of simple heuristics and scoring systems. In the development of CDRs, limitations in analytic methods and concerns with usability have generally constrained models to a preselected small set of variables judged to be clinically relevant and to rules that are easily calculated. Furthermore, CDRs frequently suffer from questions of generalizability, take years to develop, and lack the ability to be updated as new information becomes available. Newer analytic and machine learning techniques capable of harnessing the large number of variables that are already available through electronic health records (EHRs) may better predict patient outcomes and facilitate automation and deployment within clinical decision support systems. In this proof-of-concept study, a local, big data-driven, machine learning approach is compared to existing CDRs and traditional analytic methods using the prediction of sepsis in-hospital mortality as the use case. This was a retrospective study of adult ED visits admitted to the hospital meeting criteria for sepsis from October 2013 to October 2014. Sepsis was defined as meeting criteria for systemic inflammatory response syndrome with an infectious admitting diagnosis in the ED. ED visits were randomly partitioned into an 80%/20% split for training and validation. A random forest model (machine learning approach) was constructed using over 500 clinical variables from data available within the EHRs of four hospitals to predict in-hospital mortality. The machine learning prediction model was then compared to a classification and regression tree (CART) model, logistic regression model, and previously developed prediction tools on the validation data set using area under the receiver operating characteristic curve (AUC) and chi-square statistics. There were 5,278 visits among 4,676 unique patients who met criteria for sepsis. Of the 4,222 patients in the training group, 210 (5.0%) died during hospitalization, and of the 1,056 patients in the validation group, 50 (4.7%) died during hospitalization. The AUCs with 95% confidence intervals (CIs) for the different models were as follows: random forest model, 0.86 (95% CI = 0.82 to 0.90); CART model, 0.69 (95% CI = 0.62 to 0.77); logistic regression model, 0.76 (95% CI = 0.69 to 0.82); CURB-65, 0.73 (95% CI = 0.67 to 0.80); MEDS, 0.71 (95% CI = 0.63 to 0.77); and mREMS, 0.72 (95% CI = 0.65 to 0.79). The random forest model AUC was statistically different from all other models (p ≤ 0.003 for all comparisons). In this proof-of-concept study, a local big data-driven, machine learning approach outperformed existing CDRs as well as traditional analytic techniques for predicting in-hospital mortality of ED patients with sepsis. Future research should prospectively evaluate the effectiveness of this approach and whether it translates into improved clinical outcomes for high-risk sepsis patients. The methods developed serve as an example of a new model for predictive analytics in emergency care that can be automated, applied to other clinical outcomes of interest, and deployed in EHRs to enable locally relevant clinical predictions. © 2015 by the Society for Academic Emergency Medicine.

Prediction of In-hospital Mortality in Emergency Department Patients With Sepsis: A Local Big Data–Driven, Machine Learning Approach

PubMed Central

Taylor, R. Andrew; Pare, Joseph R.; Venkatesh, Arjun K.; Mowafi, Hani; Melnick, Edward R.; Fleischman, William; Hall, M. Kennedy

2018-01-01

Objectives Predictive analytics in emergency care has mostly been limited to the use of clinical decision rules (CDRs) in the form of simple heuristics and scoring systems. In the development of CDRs, limitations in analytic methods and concerns with usability have generally constrained models to a preselected small set of variables judged to be clinically relevant and to rules that are easily calculated. Furthermore, CDRs frequently suffer from questions of generalizability, take years to develop, and lack the ability to be updated as new information becomes available. Newer analytic and machine learning techniques capable of harnessing the large number of variables that are already available through electronic health records (EHRs) may better predict patient outcomes and facilitate automation and deployment within clinical decision support systems. In this proof-of-concept study, a local, big data–driven, machine learning approach is compared to existing CDRs and traditional analytic methods using the prediction of sepsis in-hospital mortality as the use case. Methods This was a retrospective study of adult ED visits admitted to the hospital meeting criteria for sepsis from October 2013 to October 2014. Sepsis was defined as meeting criteria for systemic inflammatory response syndrome with an infectious admitting diagnosis in the ED. ED visits were randomly partitioned into an 80%/20% split for training and validation. A random forest model (machine learning approach) was constructed using over 500 clinical variables from data available within the EHRs of four hospitals to predict in-hospital mortality. The machine learning prediction model was then compared to a classification and regression tree (CART) model, logistic regression model, and previously developed prediction tools on the validation data set using area under the receiver operating characteristic curve (AUC) and chi-square statistics. Results There were 5,278 visits among 4,676 unique patients who met criteria for sepsis. Of the 4,222 patients in the training group, 210 (5.0%) died during hospitalization, and of the 1,056 patients in the validation group, 50 (4.7%) died during hospitalization. The AUCs with 95% confidence intervals (CIs) for the different models were as follows: random forest model, 0.86 (95% CI = 0.82 to 0.90); CART model, 0.69 (95% CI = 0.62 to 0.77); logistic regression model, 0.76 (95% CI = 0.69 to 0.82); CURB-65, 0.73 (95% CI = 0.67 to 0.80); MEDS, 0.71 (95% CI = 0.63 to 0.77); and mREMS, 0.72 (95% CI = 0.65 to 0.79). The random forest model AUC was statistically different from all other models (p ≤ 0.003 for all comparisons). Conclusions In this proof-of-concept study, a local big data–driven, machine learning approach outperformed existing CDRs as well as traditional analytic techniques for predicting in-hospital mortality of ED patients with sepsis. Future research should prospectively evaluate the effectiveness of this approach and whether it translates into improved clinical outcomes for high-risk sepsis patients. The methods developed serve as an example of a new model for predictive analytics in emergency care that can be automated, applied to other clinical outcomes of interest, and deployed in EHRs to enable locally relevant clinical predictions. PMID:26679719

Antibacterial Activity of Imidazolium-Based Ionic Liquids Investigated by QSAR Modeling and Experimental Studies.

PubMed

Hodyna, Diana; Kovalishyn, Vasyl; Rogalsky, Sergiy; Blagodatnyi, Volodymyr; Petko, Kirill; Metelytsia, Larisa

2016-09-01

Predictive QSAR models for the inhibitors of B. subtilis and Ps. aeruginosa among imidazolium-based ionic liquids were developed using literary data. The regression QSAR models were created through Artificial Neural Network and k-nearest neighbor procedures. The classification QSAR models were constructed using WEKA-RF (random forest) method. The predictive ability of the models was tested by fivefold cross-validation; giving q(2) = 0.77-0.92 for regression models and accuracy 83-88% for classification models. Twenty synthesized samples of 1,3-dialkylimidazolium ionic liquids with predictive value of activity level of antimicrobial potential were evaluated. For all asymmetric 1,3-dialkylimidazolium ionic liquids, only compounds containing at least one radical with alkyl chain length of 12 carbon atoms showed high antibacterial activity. However, the activity of symmetric 1,3-dialkylimidazolium salts was found to have opposite relationship with the length of aliphatic radical being maximum for compounds based on 1,3-dioctylimidazolium cation. The obtained experimental results suggested that the application of classification QSAR models is more accurate for the prediction of activity of new imidazolium-based ILs as potential antibacterials. © 2016 John Wiley & Sons A/S.

Impact of Health Research Systems on Under-5 Mortality Rate: A Trend Analysis.

PubMed

Yazdizadeh, Bahareh; Parsaeian, Mahboubeh; Majdzadeh, Reza; Nikooee, Sima

2016-11-26

Between 1990 and 2015, under-5 mortality rate (U5MR) declined by 53%, from an estimated rate of 91 deaths per 1000 live births to 43, globally. The aim of this study was to determine the share of health research systems in this decrease alongside other influential factors. We used random effect regression models including the 'random intercept' and 'random intercept and random slope' models to analyze the panel data from 1990 to 2010. We selected the countries with U5MRs falling between the first and third quartiles in 1990. We used both the total articles (TA) and the number of child-specific articles (CSA) as a proxy of the health research system. In order to account for the impact of other factors, measles vaccination coverage (MVC) (as a proxy of health system performance), gross domestic product (GDP), human development index (HDI), and corruption perception index (CPI) (as proxies of development), were embedded in the model. Among all the models, 'the random intercept and random slope models' had lower residuals. The same variables of CSA, HDI, and time were significant and the coefficient of CSA was estimated at -0.17; meaning, with the addition of every 100 CSA, the rate of U5MR decreased by 17 per 1000 live births. Although the number of CSA has contributed to the reduction of U5MR, the amount of its contribution is negligible compared to the countries' development. We recommend entering different types of researches into the model separately in future research and including the variable of 'exchange between knowledge generator and user.' © 2017 The Author(s); Published by Kerman University of Medical Sciences. This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Dose-Dependent Effects of Statins for Patients with Aneurysmal Subarachnoid Hemorrhage: Meta-Regression Analysis.

PubMed

To, Minh-Son; Prakash, Shivesh; Poonnoose, Santosh I; Bihari, Shailesh

2018-05-01

The study uses meta-regression analysis to quantify the dose-dependent effects of statin pharmacotherapy on vasospasm, delayed ischemic neurologic deficits (DIND), and mortality in aneurysmal subarachnoid hemorrhage. Prospective, retrospective observational studies, and randomized controlled trials (RCTs) were retrieved by a systematic database search. Summary estimates were expressed as absolute risk (AR) for a given statin dose or control (placebo). Meta-regression using inverse variance weighting and robust variance estimation was performed to assess the effect of statin dose on transformed AR in a random effects model. Dose-dependence of predicted AR with 95% confidence interval (CI) was recovered by using Miller's Freeman-Tukey inverse. The database search and study selection criteria yielded 18 studies (2594 patients) for analysis. These included 12 RCTs, 4 retrospective observational studies, and 2 prospective observational studies. Twelve studies investigated simvastatin, whereas the remaining studies investigated atorvastatin, pravastatin, or pitavastatin, with simvastatin-equivalent doses ranging from 20 to 80 mg. Meta-regression revealed dose-dependent reductions in Freeman-Tukey-transformed AR of vasospasm (slope coefficient -0.00404, 95% CI -0.00720 to -0.00087; P = 0.0321), DIND (slope coefficient -0.00316, 95% CI -0.00586 to -0.00047; P = 0.0392), and mortality (slope coefficient -0.00345, 95% CI -0.00623 to -0.00067; P = 0.0352). The present meta-regression provides weak evidence for dose-dependent reductions in vasospasm, DIND and mortality associated with acute statin use after aneurysmal subarachnoid hemorrhage. However, the analysis was limited by substantial heterogeneity among individual studies. Greater dosing strategies are a potential consideration for future RCTs. Copyright © 2018 Elsevier Inc. All rights reserved.

Multiscale measurement error models for aggregated small area health data.

PubMed

Aregay, Mehreteab; Lawson, Andrew B; Faes, Christel; Kirby, Russell S; Carroll, Rachel; Watjou, Kevin

2016-08-01

Spatial data are often aggregated from a finer (smaller) to a coarser (larger) geographical level. The process of data aggregation induces a scaling effect which smoothes the variation in the data. To address the scaling problem, multiscale models that link the convolution models at different scale levels via the shared random effect have been proposed. One of the main goals in aggregated health data is to investigate the relationship between predictors and an outcome at different geographical levels. In this paper, we extend multiscale models to examine whether a predictor effect at a finer level hold true at a coarser level. To adjust for predictor uncertainty due to aggregation, we applied measurement error models in the framework of multiscale approach. To assess the benefit of using multiscale measurement error models, we compare the performance of multiscale models with and without measurement error in both real and simulated data. We found that ignoring the measurement error in multiscale models underestimates the regression coefficient, while it overestimates the variance of the spatially structured random effect. On the other hand, accounting for the measurement error in multiscale models provides a better model fit and unbiased parameter estimates. © The Author(s) 2016.

Impact of communities, health, and emotional-related factors on smoking use: comparison of joint modeling of mean and dispersion and Bayes' hierarchical models on add health survey.

PubMed

Pu, Jie; Fang, Di; Wilson, Jeffrey R

2017-02-03

The analysis of correlated binary data is commonly addressed through the use of conditional models with random effects included in the systematic component as opposed to generalized estimating equations (GEE) models that addressed the random component. Since the joint distribution of the observations is usually unknown, the conditional distribution is a natural approach. Our objective was to compare the fit of different binary models for correlated data in Tabaco use. We advocate that the joint modeling of the mean and dispersion may be at times just as adequate. We assessed the ability of these models to account for the intraclass correlation. In so doing, we concentrated on fitting logistic regression models to address smoking behaviors. Frequentist and Bayes' hierarchical models were used to predict conditional probabilities, and the joint modeling (GLM and GAM) models were used to predict marginal probabilities. These models were fitted to National Longitudinal Study of Adolescent to Adult Health (Add Health) data for Tabaco use. We found that people were less likely to smoke if they had higher income, high school or higher education and religious. Individuals were more likely to smoke if they had abused drug or alcohol, spent more time on TV and video games, and been arrested. Moreover, individuals who drank alcohol early in life were more likely to be a regular smoker. Children who experienced mistreatment from their parents were more likely to use Tabaco regularly. The joint modeling of the mean and dispersion models offered a flexible and meaningful method of addressing the intraclass correlation. They do not require one to identify random effects nor distinguish from one level of the hierarchy to the other. Moreover, once one can identify the significant random effects, one can obtain similar results to the random coefficient models. We found that the set of marginal models accounting for extravariation through the additional dispersion submodel produced similar results with regards to inferences and predictions. Moreover, both marginal and conditional models demonstrated similar predictive power.

Machine learning approaches to the social determinants of health in the health and retirement study.

PubMed

Seligman, Benjamin; Tuljapurkar, Shripad; Rehkopf, David

2018-04-01

Social and economic factors are important predictors of health and of recognized importance for health systems. However, machine learning, used elsewhere in the biomedical literature, has not been extensively applied to study relationships between society and health. We investigate how machine learning may add to our understanding of social determinants of health using data from the Health and Retirement Study. A linear regression of age and gender, and a parsimonious theory-based regression additionally incorporating income, wealth, and education, were used to predict systolic blood pressure, body mass index, waist circumference, and telomere length. Prediction, fit, and interpretability were compared across four machine learning methods: linear regression, penalized regressions, random forests, and neural networks. All models had poor out-of-sample prediction. Most machine learning models performed similarly to the simpler models. However, neural networks greatly outperformed the three other methods. Neural networks also had good fit to the data ( R 2 between 0.4-0.6, versus <0.3 for all others). Across machine learning models, nine variables were frequently selected or highly weighted as predictors: dental visits, current smoking, self-rated health, serial-seven subtractions, probability of receiving an inheritance, probability of leaving an inheritance of at least $10,000, number of children ever born, African-American race, and gender. Some of the machine learning methods do not improve prediction or fit beyond simpler models, however, neural networks performed well. The predictors identified across models suggest underlying social factors that are important predictors of biological indicators of chronic disease, and that the non-linear and interactive relationships between variables fundamental to the neural network approach may be important to consider.

Comparison of Linear and Non-linear Regression Analysis to Determine Pulmonary Pressure in Hyperthyroidism.

PubMed

Scarneciu, Camelia C; Sangeorzan, Livia; Rus, Horatiu; Scarneciu, Vlad D; Varciu, Mihai S; Andreescu, Oana; Scarneciu, Ioan

2017-01-01

This study aimed at assessing the incidence of pulmonary hypertension (PH) at newly diagnosed hyperthyroid patients and at finding a simple model showing the complex functional relation between pulmonary hypertension in hyperthyroidism and the factors causing it. The 53 hyperthyroid patients (H-group) were evaluated mainly by using an echocardiographical method and compared with 35 euthyroid (E-group) and 25 healthy people (C-group). In order to identify the factors causing pulmonary hypertension the statistical method of comparing the values of arithmetical means is used. The functional relation between the two random variables (PAPs and each of the factors determining it within our research study) can be expressed by linear or non-linear function. By applying the linear regression method described by a first-degree equation the line of regression (linear model) has been determined; by applying the non-linear regression method described by a second degree equation, a parabola-type curve of regression (non-linear or polynomial model) has been determined. We made the comparison and the validation of these two models by calculating the determination coefficient (criterion 1), the comparison of residuals (criterion 2), application of AIC criterion (criterion 3) and use of F-test (criterion 4). From the H-group, 47% have pulmonary hypertension completely reversible when obtaining euthyroidism. The factors causing pulmonary hypertension were identified: previously known- level of free thyroxin, pulmonary vascular resistance, cardiac output; new factors identified in this study- pretreatment period, age, systolic blood pressure. According to the four criteria and to the clinical judgment, we consider that the polynomial model (graphically parabola- type) is better than the linear one. The better model showing the functional relation between the pulmonary hypertension in hyperthyroidism and the factors identified in this study is given by a polynomial equation of second degree where the parabola is its graphical representation.

Alternative methods to evaluate trial level surrogacy.

PubMed

Abrahantes, Josè Cortiñas; Shkedy, Ziv; Molenberghs, Geert

2008-01-01

The evaluation and validation of surrogate endpoints have been extensively studied in the last decade. Prentice [1] and Freedman, Graubard and Schatzkin [2] laid the foundations for the evaluation of surrogate endpoints in randomized clinical trials. Later, Buyse et al. [5] proposed a meta-analytic methodology, producing different methods for different settings, which was further studied by Alonso and Molenberghs [9], in their unifying approach based on information theory. In this article, we focus our attention on the trial-level surrogacy and propose alternative procedures to evaluate such surrogacy measure, which do not pre-specify the type of association. A promising correction based on cross-validation is investigated. As well as the construction of confidence intervals for this measure. In order to avoid making assumption about the type of relationship between the treatment effects and its distribution, a collection of alternative methods, based on regression trees, bagging, random forests, and support vector machines, combined with bootstrap-based confidence interval and, should one wish, in conjunction with a cross-validation based correction, will be proposed and applied. We apply the various strategies to data from three clinical studies: in opthalmology, in advanced colorectal cancer, and in schizophrenia. The results obtained for the three case studies are compared; they indicate that using random forest or bagging models produces larger estimated values for the surrogacy measure, which are in general stabler and the confidence interval narrower than linear regression and support vector regression. For the advanced colorectal cancer studies, we even found the trial-level surrogacy is considerably different from what has been reported. In general the alternative methods are more computationally demanding, and specially the calculation of the confidence intervals, require more computational time that the delta-method counterpart. First, more flexible modeling techniques can be used, allowing for other type of association. Second, when no cross-validation-based correction is applied, overly optimistic trial-level surrogacy estimates will be found, thus cross-validation is highly recommendable. Third, the use of the delta method to calculate confidence intervals is not recommendable since it makes assumptions valid only in very large samples. It may also produce range-violating limits. We therefore recommend alternatives: bootstrap methods in general. Also, the information-theoretic approach produces comparable results with the bagging and random forest approaches, when cross-validation correction is applied. It is also important to observe that, even for the case in which the linear model might be a good option too, bagging methods perform well too, and their confidence intervals were more narrow.

The value of a statistical life: a meta-analysis with a mixed effects regression model.

PubMed

Bellavance, François; Dionne, Georges; Lebeau, Martin

2009-03-01

The value of a statistical life (VSL) is a very controversial topic, but one which is essential to the optimization of governmental decisions. We see a great variability in the values obtained from different studies. The source of this variability needs to be understood, in order to offer public decision-makers better guidance in choosing a value and to set clearer guidelines for future research on the topic. This article presents a meta-analysis based on 39 observations obtained from 37 studies (from nine different countries) which all use a hedonic wage method to calculate the VSL. Our meta-analysis is innovative in that it is the first to use the mixed effects regression model [Raudenbush, S.W., 1994. Random effects models. In: Cooper, H., Hedges, L.V. (Eds.), The Handbook of Research Synthesis. Russel Sage Foundation, New York] to analyze studies on the value of a statistical life. We conclude that the variability found in the values studied stems in large part from differences in methodologies.

The prediction of food additives in the fruit juice based on electronic nose with chemometrics.

PubMed

Qiu, Shanshan; Wang, Jun

2017-09-01

Food additives are added to products to enhance their taste, and preserve flavor or appearance. While their use should be restricted to achieve a technological benefit, the contents of food additives should be also strictly controlled. In this study, E-nose was applied as an alternative to traditional monitoring technologies for determining two food additives, namely benzoic acid and chitosan. For quantitative monitoring, support vector machine (SVM), random forest (RF), extreme learning machine (ELM) and partial least squares regression (PLSR) were applied to establish regression models between E-nose signals and the amount of food additives in fruit juices. The monitoring models based on ELM and RF reached higher correlation coefficients (R 2 s) and lower root mean square errors (RMSEs) than models based on PLSR and SVM. This work indicates that E-nose combined with RF or ELM can be a cost-effective, easy-to-build and rapid detection system for food additive monitoring. Copyright © 2017 Elsevier Ltd. All rights reserved.

Latent Class Models in action: bridging social capital & Internet usage.

PubMed

Neves, Barbara Barbosa; Fonseca, Jaime R S

2015-03-01

This paper explores how Latent Class Models (LCM) can be applied in social research, when the basic assumptions of regression models cannot be validated. We examine the usefulness of this method with data collected from a study on the relationship between bridging social capital and the Internet. Social capital is defined here as the resources that are potentially available in one's social ties. Bridging is a dimension of social capital, usually related to weak ties (acquaintances), and a source of instrumental resources such as information. The study surveyed a stratified random sample of 417 inhabitants of Lisbon, Portugal. We used LCM to create the variable bridging social capital, but also to estimate the relationship between bridging social capital and Internet usage when we encountered convergence problems with the logistic regression analysis. We conclude by showing a positive relationship between bridging and Internet usage, and by discussing the potential of LCM for social science research. Copyright © 2014 Elsevier Inc. All rights reserved.

[On the effectiveness of the homeopathic remedy Arnica montana].

PubMed

Lüdtke, Rainer; Hacke, Daniela

2005-11-01

Arnica montana is a homeopathic remedy often prescribed after traumata and injuries. To assess whether Arnica is effective beyond placebo and to identify factors which support or contradict this effectiveness. All prospective, controlled trials on the effectiveness of homeopathic Arnica were included. Overall effectiveness was assessed by meta-analysis and meta-regression techniques. 68 comparisons from 49 clinical trials show a significant effectiveness of Arnica in traumatic injuries in random effects meta-analysis (odds ratio [OR], 0.36; 95% confidence interval [CI], 0.24-0.55), but not in meta-regression models (OR, 0.37; CI, 0.11-1.24). We found no evidence for publication bias. Studies from Medline-listed journals and high-quality studies are less likely to report positive results (p = 0.0006 and p = 0.0167). The hypothesis that homeopathic Arnica is effective could neither be proved nor rejected. All trials were highly heterogeneous, meta-regression does not help to explain this heterogeneity substantially.

Using structural equation modeling to construct calibration equations relating PM2.5 mass concentration samplers to the federal reference method sampler

NASA Astrophysics Data System (ADS)

Bilonick, Richard A.; Connell, Daniel P.; Talbott, Evelyn O.; Rager, Judith R.; Xue, Tao

2015-02-01

The objective of this study was to remove systematic bias among fine particulate matter (PM2.5) mass concentration measurements made by different types of samplers used in the Pittsburgh Aerosol Research and Inhalation Epidemiology Study (PARIES). PARIES is a retrospective epidemiology study that aims to provide a comprehensive analysis of the associations between air quality and human health effects in the Pittsburgh, Pennsylvania, region from 1999 to 2008. Calibration was needed in order to minimize the amount of systematic error in PM2.5 exposure estimation as a result of including data from 97 different PM2.5 samplers at 47 monitoring sites. Ordinary regression often has been used for calibrating air quality measurements from pairs of measurement devices; however, this is only appropriate when one of the two devices (the "independent" variable) is free from random error, which is rarely the case. A group of methods known as "errors-in-variables" (e.g., Deming regression, reduced major axis regression) has been developed to handle calibration between two devices when both are subject to random error, but these methods require information on the relative sizes of the random errors for each device, which typically cannot be obtained from the observed data. When data from more than two devices (or repeats of the same device) are available, the additional information is not used to inform the calibration. A more general approach that often has been overlooked is the use of a measurement error structural equation model (SEM) that allows the simultaneous comparison of three or more devices (or repeats). The theoretical underpinnings of all of these approaches to calibration are described, and the pros and cons of each are discussed. In particular, it is shown that both ordinary regression (when used for calibration) and Deming regression are particular examples of SEMs but with substantial deficiencies. To illustrate the use of SEMs, the 7865 daily average PM2.5 mass concentration measurements made by seven collocated samplers at an urban monitoring site in Pittsburgh, Pennsylvania, were used. These samplers, which included three federal reference method (FRM) samplers, three speciation samplers, and a tapered element oscillating microbalance (TEOM), operated at various times during the 10-year PARIES study period. Because TEOM measurements are known to depend on temperature, the constructed SEM provided calibration equations relating the TEOM to the FRM and speciation samplers as a function of ambient temperature. It was shown that TEOM imprecision and TEOM bias (relative to the FRM) both decreased as temperature increased. It also was shown that the temperature dependency for bias was non-linear and followed a sigmoidal (logistic) pattern. The speciation samplers exhibited only small bias relative to the FRM samplers, although the FRM samplers were shown to be substantially more precise than both the TEOM and the speciation samplers. Comparison of the SEM results to pairwise simple linear regression results showed that the regression results can differ substantially from the correctly-derived calibration equations, especially if the less-precise device is used as the independent variable in the regression.

A hybrid training approach for leaf area index estimation via Cubist and random forests machine-learning

NASA Astrophysics Data System (ADS)

Houborg, Rasmus; McCabe, Matthew F.

2018-01-01

With an increasing volume and dimensionality of Earth observation data, enhanced integration of machine-learning methodologies is needed to effectively analyze and utilize these information rich datasets. In machine-learning, a training dataset is required to establish explicit associations between a suite of explanatory 'predictor' variables and the target property. The specifics of this learning process can significantly influence model validity and portability, with a higher generalization level expected with an increasing number of observable conditions being reflected in the training dataset. Here we propose a hybrid training approach for leaf area index (LAI) estimation, which harnesses synergistic attributes of scattered in-situ measurements and systematically distributed physically based model inversion results to enhance the information content and spatial representativeness of the training data. To do this, a complimentary training dataset of independent LAI was derived from a regularized model inversion of RapidEye surface reflectances and subsequently used to guide the development of LAI regression models via Cubist and random forests (RF) decision tree methods. The application of the hybrid training approach to a broad set of Landsat 8 vegetation index (VI) predictor variables resulted in significantly improved LAI prediction accuracies and spatial consistencies, relative to results relying on in-situ measurements alone for model training. In comparing the prediction capacity and portability of the two machine-learning algorithms, a pair of relatively simple multi-variate regression models established by Cubist performed best, with an overall relative mean absolute deviation (rMAD) of ∼11%, determined based on a stringent scene-specific cross-validation approach. In comparison, the portability of RF regression models was less effective (i.e., an overall rMAD of ∼15%), which was attributed partly to model saturation at high LAI in association with inherent extrapolation and transferability limitations. Explanatory VIs formed from bands in the near-infrared (NIR) and shortwave infrared domains (e.g., NDWI) were associated with the highest predictive ability, whereas Cubist models relying entirely on VIs based on NIR and red band combinations (e.g., NDVI) were associated with comparatively high uncertainties (i.e., rMAD ∼ 21%). The most transferable and best performing models were based on combinations of several predictor variables, which included both NDWI- and NDVI-like variables. In this process, prior screening of input VIs based on an assessment of variable relevance served as an effective mechanism for optimizing prediction accuracies from both Cubist and RF. While this study demonstrated benefit in combining data mining operations with physically based constraints via a hybrid training approach, the concept of transferability and portability warrants further investigations in order to realize the full potential of emerging machine-learning techniques for regression purposes.

Improved predictive mapping of indoor radon concentrations using ensemble regression trees based on automatic clustering of geological units.

PubMed

Kropat, Georg; Bochud, Francois; Jaboyedoff, Michel; Laedermann, Jean-Pascal; Murith, Christophe; Palacios Gruson, Martha; Baechler, Sébastien

2015-09-01

According to estimations around 230 people die as a result of radon exposure in Switzerland. This public health concern makes reliable indoor radon prediction and mapping methods necessary in order to improve risk communication to the public. The aim of this study was to develop an automated method to classify lithological units according to their radon characteristics and to develop mapping and predictive tools in order to improve local radon prediction. About 240 000 indoor radon concentration (IRC) measurements in about 150 000 buildings were available for our analysis. The automated classification of lithological units was based on k-medoids clustering via pair-wise Kolmogorov distances between IRC distributions of lithological units. For IRC mapping and prediction we used random forests and Bayesian additive regression trees (BART). The automated classification groups lithological units well in terms of their IRC characteristics. Especially the IRC differences in metamorphic rocks like gneiss are well revealed by this method. The maps produced by random forests soundly represent the regional difference of IRCs in Switzerland and improve the spatial detail compared to existing approaches. We could explain 33% of the variations in IRC data with random forests. Additionally, the influence of a variable evaluated by random forests shows that building characteristics are less important predictors for IRCs than spatial/geological influences. BART could explain 29% of IRC variability and produced maps that indicate the prediction uncertainty. Ensemble regression trees are a powerful tool to model and understand the multidimensional influences on IRCs. Automatic clustering of lithological units complements this method by facilitating the interpretation of radon properties of rock types. This study provides an important element for radon risk communication. Future approaches should consider taking into account further variables like soil gas radon measurements as well as more detailed geological information. Copyright © 2015 Elsevier Ltd. All rights reserved.

Safety and efficacy of antihypertensive prescription at emergency department discharge.

PubMed

Brody, Aaron; Rahman, Tahsin; Reed, Brian; Millis, Scott; Ference, Brian; Flack, John M; Levy, Phillip D

2015-05-01

Poor blood pressure (BP) control is a primary risk factor for target organ damage in the heart, brain, and kidney. Uncontrolled hypertension is common among emergency department (ED) patients, particularly in underresourced settings, but it is unclear what role ED providers should play in the management of chronic antihypertensive therapy. The objective was to evaluate the safety and efficacy of prescribing antihypertensive therapy from the ED. This was a retrospective study of data pooled from two prospective, longitudinal, randomized controlled trials, both of which enrolled ED patients with asymptomatic hypertension. Antihypertensives were prescribed at emergency physician discretion, and this was not related to randomization arm. Demographic data, BP at screening and randomization visit, and data on adverse effects potentially related to antihypertensive therapy were compiled. Means were compared using Student's t-tests, and proportions were compared using chi-square tests. The effect of antihypertensive therapy on BP control was further analyzed using multivariable regression modeling controlling for age, race, sex, hypertension history, study cohort, and ED BP. Data were abstracted for 217 subjects. The median interval from ED visit to randomization was 12 days. Seventy-six subjects (35%) received one or more prescriptions for antihypertensive therapy. Age, sex, race, hypertension history, and mean duration of hypertension were equivalent between groups. Although mean ED BP was higher among those who received prescriptions, the mean systolic BP (sBP) reduction from ED to randomization was significantly greater (difference = 19 mm Hg, 95% confidence interval = 12 to 26 mm Hg). No patient in either group had an sBP less than 100 mm Hg at randomization. On multiple regression modeling, randomization sBP reduction was independently associated with antihypertensive prescription (p = 0.001). The incidence of adverse effects was equivalent and low in both groups. No new neurological deficits, ischemic events, or life-threatening anaphylactic reactions were reported in either group. Prescription of antihypertensive medication from the ED is associated with significantly lower sBP at short-term outpatient follow-up. Antihypertensive therapy was not associated with an increased incidence of adverse events, and BP reduction did not exceed potentially harmful levels. Initiation of chronic antihypertensive therapy in the ED is safe and effective and may be a reasonable consideration for at-risk populations. © 2015 by the Society for Academic Emergency Medicine.

surrosurv: An R package for the evaluation of failure time surrogate endpoints in individual patient data meta-analyses of randomized clinical trials.

PubMed

Rotolo, Federico; Paoletti, Xavier; Michiels, Stefan

2018-03-01

Surrogate endpoints are attractive for use in clinical trials instead of well-established endpoints because of practical convenience. To validate a surrogate endpoint, two important measures can be estimated in a meta-analytic context when individual patient data are available: the R indiv 2 or the Kendall's τ at the individual level, and the R trial 2 at the trial level. We aimed at providing an R implementation of classical and well-established as well as more recent statistical methods for surrogacy assessment with failure time endpoints. We also intended incorporating utilities for model checking and visualization and data generating methods described in the literature to date. In the case of failure time endpoints, the classical approach is based on two steps. First, a Kendall's τ is estimated as measure of individual level surrogacy using a copula model. Then, the R trial 2 is computed via a linear regression of the estimated treatment effects; at this second step, the estimation uncertainty can be accounted for via measurement-error model or via weights. In addition to the classical approach, we recently developed an approach based on bivariate auxiliary Poisson models with individual random effects to measure the Kendall's τ and treatment-by-trial interactions to measure the R trial 2 . The most common data simulation models described in the literature are based on: copula models, mixed proportional hazard models, and mixture of half-normal and exponential random variables. The R package surrosurv implements the classical two-step method with Clayton, Plackett, and Hougaard copulas. It also allows to optionally adjusting the second-step linear regression for measurement-error. The mixed Poisson approach is implemented with different reduced models in addition to the full model. We present the package functions for estimating the surrogacy models, for checking their convergence, for performing leave-one-trial-out cross-validation, and for plotting the results. We illustrate their use in practice on individual patient data from a meta-analysis of 4069 patients with advanced gastric cancer from 20 trials of chemotherapy. The surrosurv package provides an R implementation of classical and recent statistical methods for surrogacy assessment of failure time endpoints. Flexible simulation functions are available to generate data according to the methods described in the literature. Copyright © 2017 Elsevier B.V. All rights reserved.

Hyperparameterization of soil moisture statistical models for North America with Ensemble Learning Models (Elm)

NASA Astrophysics Data System (ADS)

Steinberg, P. D.; Brener, G.; Duffy, D.; Nearing, G. S.; Pelissier, C.

2017-12-01

Hyperparameterization, of statistical models, i.e. automated model scoring and selection, such as evolutionary algorithms, grid searches, and randomized searches, can improve forecast model skill by reducing errors associated with model parameterization, model structure, and statistical properties of training data. Ensemble Learning Models (Elm), and the related Earthio package, provide a flexible interface for automating the selection of parameters and model structure for machine learning models common in climate science and land cover classification, offering convenient tools for loading NetCDF, HDF, Grib, or GeoTiff files, decomposition methods like PCA and manifold learning, and parallel training and prediction with unsupervised and supervised classification, clustering, and regression estimators. Continuum Analytics is using Elm to experiment with statistical soil moisture forecasting based on meteorological forcing data from NASA's North American Land Data Assimilation System (NLDAS). There Elm is using the NSGA-2 multiobjective optimization algorithm for optimizing statistical preprocessing of forcing data to improve goodness-of-fit for statistical models (i.e. feature engineering). This presentation will discuss Elm and its components, including dask (distributed task scheduling), xarray (data structures for n-dimensional arrays), and scikit-learn (statistical preprocessing, clustering, classification, regression), and it will show how NSGA-2 is being used for automate selection of soil moisture forecast statistical models for North America.

Spatial distribution of psychotic disorders in an urban area of France: an ecological study.

PubMed

Pignon, Baptiste; Schürhoff, Franck; Baudin, Grégoire; Ferchiou, Aziz; Richard, Jean-Romain; Saba, Ghassen; Leboyer, Marion; Kirkbride, James B; Szöke, Andrei

2016-05-18

Previous analyses of neighbourhood variations of non-affective psychotic disorders (NAPD) have focused mainly on incidence. However, prevalence studies provide important insights on factors associated with disease evolution as well as for healthcare resource allocation. This study aimed to investigate the distribution of prevalent NAPD cases in an urban area in France. The number of cases in each neighbourhood was modelled as a function of potential confounders and ecological variables, namely: migrant density, economic deprivation and social fragmentation. This was modelled using statistical models of increasing complexity: frequentist models (using Poisson and negative binomial regressions), and several Bayesian models. For each model, assumptions validity were checked and compared as to how this fitted to the data, in order to test for possible spatial variation in prevalence. Data showed significant overdispersion (invalidating the Poisson regression model) and residual autocorrelation (suggesting the need to use Bayesian models). The best Bayesian model was Leroux's model (i.e. a model with both strong correlation between neighbouring areas and weaker correlation between areas further apart), with economic deprivation as an explanatory variable (OR = 1.13, 95% CI [1.02-1.25]). In comparison with frequentist methods, the Bayesian model showed a better fit. The number of cases showed non-random spatial distribution and was linked to economic deprivation.