Parametric regression model for survival data: Weibull regression model as an example

PubMed Central

2016-01-01

Weibull regression model is one of the most popular forms of parametric regression model that it provides estimate of baseline hazard function, as well as coefficients for covariates. Because of technical difficulties, Weibull regression model is seldom used in medical literature as compared to the semi-parametric proportional hazard model. To make clinical investigators familiar with Weibull regression model, this article introduces some basic knowledge on Weibull regression model and then illustrates how to fit the model with R software. The SurvRegCensCov package is useful in converting estimated coefficients to clinical relevant statistics such as hazard ratio (HR) and event time ratio (ETR). Model adequacy can be assessed by inspecting Kaplan-Meier curves stratified by categorical variable. The eha package provides an alternative method to model Weibull regression model. The check.dist() function helps to assess goodness-of-fit of the model. Variable selection is based on the importance of a covariate, which can be tested using anova() function. Alternatively, backward elimination starting from a full model is an efficient way for model development. Visualization of Weibull regression model after model development is interesting that it provides another way to report your findings. PMID:28149846

Analyzing hospitalization data: potential limitations of Poisson regression.

PubMed

Weaver, Colin G; Ravani, Pietro; Oliver, Matthew J; Austin, Peter C; Quinn, Robert R

2015-08-01

Poisson regression is commonly used to analyze hospitalization data when outcomes are expressed as counts (e.g. number of days in hospital). However, data often violate the assumptions on which Poisson regression is based. More appropriate extensions of this model, while available, are rarely used. We compared hospitalization data between 206 patients treated with hemodialysis (HD) and 107 treated with peritoneal dialysis (PD) using Poisson regression and compared results from standard Poisson regression with those obtained using three other approaches for modeling count data: negative binomial (NB) regression, zero-inflated Poisson (ZIP) regression and zero-inflated negative binomial (ZINB) regression. We examined the appropriateness of each model and compared the results obtained with each approach. During a mean 1.9 years of follow-up, 183 of 313 patients (58%) were never hospitalized (indicating an excess of 'zeros'). The data also displayed overdispersion (variance greater than mean), violating another assumption of the Poisson model. Using four criteria, we determined that the NB and ZINB models performed best. According to these two models, patients treated with HD experienced similar hospitalization rates as those receiving PD {NB rate ratio (RR): 1.04 [bootstrapped 95% confidence interval (CI): 0.49-2.20]; ZINB summary RR: 1.21 (bootstrapped 95% CI 0.60-2.46)}. Poisson and ZIP models fit the data poorly and had much larger point estimates than the NB and ZINB models [Poisson RR: 1.93 (bootstrapped 95% CI 0.88-4.23); ZIP summary RR: 1.84 (bootstrapped 95% CI 0.88-3.84)]. We found substantially different results when modeling hospitalization data, depending on the approach used. Our results argue strongly for a sound model selection process and improved reporting around statistical methods used for modeling count data. © The Author 2015. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.

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.

Prediction of unwanted pregnancies using logistic regression, probit regression and discriminant analysis

PubMed Central

Ebrahimzadeh, Farzad; Hajizadeh, Ebrahim; Vahabi, Nasim; Almasian, Mohammad; Bakhteyar, Katayoon

2015-01-01

Background: Unwanted pregnancy not intended by at least one of the parents has undesirable consequences for the family and the society. In the present study, three classification models were used and compared to predict unwanted pregnancies in an urban population. Methods: In this cross-sectional study, 887 pregnant mothers referring to health centers in Khorramabad, Iran, in 2012 were selected by the stratified and cluster sampling; relevant variables were measured and for prediction of unwanted pregnancy, logistic regression, discriminant analysis, and probit regression models and SPSS software version 21 were used. To compare these models, indicators such as sensitivity, specificity, the area under the ROC curve, and the percentage of correct predictions were used. Results: The prevalence of unwanted pregnancies was 25.3%. The logistic and probit regression models indicated that parity and pregnancy spacing, contraceptive methods, household income and number of living male children were related to unwanted pregnancy. The performance of the models based on the area under the ROC curve was 0.735, 0.733, and 0.680 for logistic regression, probit regression, and linear discriminant analysis, respectively. Conclusion: Given the relatively high prevalence of unwanted pregnancies in Khorramabad, it seems necessary to revise family planning programs. Despite the similar accuracy of the models, if the researcher is interested in the interpretability of the results, the use of the logistic regression model is recommended. PMID:26793655

Prediction of unwanted pregnancies using logistic regression, probit regression and discriminant analysis.

PubMed

Ebrahimzadeh, Farzad; Hajizadeh, Ebrahim; Vahabi, Nasim; Almasian, Mohammad; Bakhteyar, Katayoon

2015-01-01

Unwanted pregnancy not intended by at least one of the parents has undesirable consequences for the family and the society. In the present study, three classification models were used and compared to predict unwanted pregnancies in an urban population. In this cross-sectional study, 887 pregnant mothers referring to health centers in Khorramabad, Iran, in 2012 were selected by the stratified and cluster sampling; relevant variables were measured and for prediction of unwanted pregnancy, logistic regression, discriminant analysis, and probit regression models and SPSS software version 21 were used. To compare these models, indicators such as sensitivity, specificity, the area under the ROC curve, and the percentage of correct predictions were used. The prevalence of unwanted pregnancies was 25.3%. The logistic and probit regression models indicated that parity and pregnancy spacing, contraceptive methods, household income and number of living male children were related to unwanted pregnancy. The performance of the models based on the area under the ROC curve was 0.735, 0.733, and 0.680 for logistic regression, probit regression, and linear discriminant analysis, respectively. Given the relatively high prevalence of unwanted pregnancies in Khorramabad, it seems necessary to revise family planning programs. Despite the similar accuracy of the models, if the researcher is interested in the interpretability of the results, the use of the logistic regression model is recommended.

A comparative study between nonlinear regression and nonparametric approaches for modelling Phalaris paradoxa seedling emergence

USDA-ARS?s Scientific Manuscript database

Parametric non-linear regression (PNR) techniques commonly are used to develop weed seedling emergence models. Such techniques, however, require statistical assumptions that are difficult to meet. To examine and overcome these limitations, we compared PNR with a nonparametric estimation technique. F...

Application of third molar development and eruption models in estimating dental age in Malay sub-adults.

PubMed

Mohd Yusof, Mohd Yusmiaidil Putera; Cauwels, Rita; Deschepper, Ellen; Martens, Luc

2015-08-01

The third molar development (TMD) has been widely utilized as one of the radiographic method for dental age estimation. By using the same radiograph of the same individual, third molar eruption (TME) information can be incorporated to the TMD regression model. This study aims to evaluate the performance of dental age estimation in individual method models and the combined model (TMD and TME) based on the classic regressions of multiple linear and principal component analysis. A sample of 705 digital panoramic radiographs of Malay sub-adults aged between 14.1 and 23.8 years was collected. The techniques described by Gleiser and Hunt (modified by Kohler) and Olze were employed to stage the TMD and TME, respectively. The data was divided to develop three respective models based on the two regressions of multiple linear and principal component analysis. The trained models were then validated on the test sample and the accuracy of age prediction was compared between each model. The coefficient of determination (R²) and root mean square error (RMSE) were calculated. In both genders, adjusted R² yielded an increment in the linear regressions of combined model as compared to the individual models. The overall decrease in RMSE was detected in combined model as compared to TMD (0.03-0.06) and TME (0.2-0.8). In principal component regression, low value of adjusted R(2) and high RMSE except in male were exhibited in combined model. Dental age estimation is better predicted using combined model in multiple linear regression models. Copyright © 2015 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved.

Estimation of Standard Error of Regression Effects in Latent Regression Models Using Binder's Linearization. Research Report. ETS RR-07-09

ERIC Educational Resources Information Center

Li, Deping; Oranje, Andreas

2007-01-01

Two versions of a general method for approximating standard error of regression effect estimates within an IRT-based latent regression model are compared. The general method is based on Binder's (1983) approach, accounting for complex samples and finite populations by Taylor series linearization. In contrast, the current National Assessment of…

Applicability of Monte Carlo cross validation technique for model development and validation using generalised least squares regression

NASA Astrophysics Data System (ADS)

Haddad, Khaled; Rahman, Ataur; A Zaman, Mohammad; Shrestha, Surendra

2013-03-01

SummaryIn regional hydrologic regression analysis, model selection and validation are regarded as important steps. Here, the model selection is usually based on some measurements of goodness-of-fit between the model prediction and observed data. In Regional Flood Frequency Analysis (RFFA), leave-one-out (LOO) validation or a fixed percentage leave out validation (e.g., 10%) is commonly adopted to assess the predictive ability of regression-based prediction equations. This paper develops a Monte Carlo Cross Validation (MCCV) technique (which has widely been adopted in Chemometrics and Econometrics) in RFFA using Generalised Least Squares Regression (GLSR) and compares it with the most commonly adopted LOO validation approach. The study uses simulated and regional flood data from the state of New South Wales in Australia. It is found that when developing hydrologic regression models, application of the MCCV is likely to result in a more parsimonious model than the LOO. It has also been found that the MCCV can provide a more realistic estimate of a model's predictive ability when compared with the LOO.

Regression analysis and transfer function in estimating the parameters of central pulse waves from brachial pulse wave.

PubMed

Chai Rui; Li Si-Man; Xu Li-Sheng; Yao Yang; Hao Li-Ling

2017-07-01

This study mainly analyzed the parameters such as ascending branch slope (A_slope), dicrotic notch height (Hn), diastolic area (Ad) and systolic area (As) diastolic blood pressure (DBP), systolic blood pressure (SBP), pulse pressure (PP), subendocardial viability ratio (SEVR), waveform parameter (k), stroke volume (SV), cardiac output (CO) and peripheral resistance (RS) of central pulse wave invasively and non-invasively measured. These parameters extracted from the central pulse wave invasively measured were compared with the parameters measured from the brachial pulse waves by a regression model and a transfer function model. The accuracy of the parameters which were estimated by the regression model and the transfer function model was compared too. Our findings showed that in addition to the k value, the above parameters of the central pulse wave and the brachial pulse wave invasively measured had positive correlation. Both the regression model parameters including A_slope, DBP, SEVR and the transfer function model parameters had good consistency with the parameters invasively measured, and they had the same effect of consistency. The regression equations of the three parameters were expressed by Y'=a+bx. The SBP, PP, SV, CO of central pulse wave could be calculated through the regression model, but their accuracies were worse than that of transfer function model.

Comparative study of biodegradability prediction of chemicals using decision trees, functional trees, and logistic regression.

PubMed

Chen, Guangchao; Li, Xuehua; Chen, Jingwen; Zhang, Ya-Nan; Peijnenburg, Willie J G M

2014-12-01

Biodegradation is the principal environmental dissipation process of chemicals. As such, it is a dominant factor determining the persistence and fate of organic chemicals in the environment, and is therefore of critical importance to chemical management and regulation. In the present study, the authors developed in silico methods assessing biodegradability based on a large heterogeneous set of 825 organic compounds, using the techniques of the C4.5 decision tree, the functional inner regression tree, and logistic regression. External validation was subsequently carried out by 2 independent test sets of 777 and 27 chemicals. As a result, the functional inner regression tree exhibited the best predictability with predictive accuracies of 81.5% and 81.0%, respectively, on the training set (825 chemicals) and test set I (777 chemicals). Performance of the developed models on the 2 test sets was subsequently compared with that of the Estimation Program Interface (EPI) Suite Biowin 5 and Biowin 6 models, which also showed a better predictability of the functional inner regression tree model. The model built in the present study exhibits a reasonable predictability compared with existing models while possessing a transparent algorithm. Interpretation of the mechanisms of biodegradation was also carried out based on the models developed. © 2014 SETAC.

Estimating linear temporal trends from aggregated environmental monitoring data

USGS Publications Warehouse

Erickson, Richard A.; Gray, Brian R.; Eager, Eric A.

2017-01-01

Trend estimates are often used as part of environmental monitoring programs. These trends inform managers (e.g., are desired species increasing or undesired species decreasing?). Data collected from environmental monitoring programs is often aggregated (i.e., averaged), which confounds sampling and process variation. State-space models allow sampling variation and process variations to be separated. We used simulated time-series to compare linear trend estimations from three state-space models, a simple linear regression model, and an auto-regressive model. We also compared the performance of these five models to estimate trends from a long term monitoring program. We specifically estimated trends for two species of fish and four species of aquatic vegetation from the Upper Mississippi River system. We found that the simple linear regression had the best performance of all the given models because it was best able to recover parameters and had consistent numerical convergence. Conversely, the simple linear regression did the worst job estimating populations in a given year. The state-space models did not estimate trends well, but estimated population sizes best when the models converged. We found that a simple linear regression performed better than more complex autoregression and state-space models when used to analyze aggregated environmental monitoring data.

The role of empathy and emotional intelligence in nurses' communication attitudes using regression models and fuzzy-set qualitative comparative analysis models.

PubMed

Giménez-Espert, María Del Carmen; Prado-Gascó, Vicente Javier

2018-03-01

To analyse link between empathy and emotional intelligence as a predictor of nurses' attitudes towards communication while comparing the contribution of emotional aspects and attitudinal elements on potential behaviour. Nurses' attitudes towards communication, empathy and emotional intelligence are key skills for nurses involved in patient care. There are currently no studies analysing this link, and its investigation is needed because attitudes may influence communication behaviours. Correlational study. To attain this goal, self-reported instruments (attitudes towards communication of nurses, trait emotional intelligence (Trait Emotional Meta-Mood Scale) and Jefferson Scale of Nursing Empathy (Jefferson Scale Nursing Empathy) were collected from 460 nurses between September 2015-February 2016. Two different analytical methodologies were used: traditional regression models and fuzzy-set qualitative comparative analysis models. The results of the regression model suggest that cognitive dimensions of attitude are a significant and positive predictor of the behavioural dimension. The perspective-taking dimension of empathy and the emotional-clarity dimension of emotional intelligence were significant positive predictors of the dimensions of attitudes towards communication, except for the affective dimension (for which the association was negative). The results of the fuzzy-set qualitative comparative analysis models confirm that the combination of high levels of cognitive dimension of attitudes, perspective-taking and emotional clarity explained high levels of the behavioural dimension of attitude. Empathy and emotional intelligence are predictors of nurses' attitudes towards communication, and the cognitive dimension of attitude is a good predictor of the behavioural dimension of attitudes towards communication of nurses in both regression models and fuzzy-set qualitative comparative analysis. In general, the fuzzy-set qualitative comparative analysis models appear to be better predictors than the regression models are. To evaluate current practices, establish intervention strategies and evaluate their effectiveness. The evaluation of these variables and their relationships are important in creating a satisfied and sustainable workforce and improving quality of care and patient health. © 2018 John Wiley & Sons Ltd.

Incremental Net Effects in Multiple Regression

ERIC Educational Resources Information Center

Lipovetsky, Stan; Conklin, Michael

2005-01-01

A regular problem in regression analysis is estimating the comparative importance of the predictors in the model. This work considers the 'net effects', or shares of the predictors in the coefficient of the multiple determination, which is a widely used characteristic of the quality of a regression model. Estimation of the net effects can be a…

Climate variations and salmonellosis transmission in Adelaide, South Australia: a comparison between regression models

NASA Astrophysics Data System (ADS)

Zhang, Ying; Bi, Peng; Hiller, Janet

2008-01-01

This is the first study to identify appropriate regression models for the association between climate variation and salmonellosis transmission. A comparison between different regression models was conducted using surveillance data in Adelaide, South Australia. By using notified salmonellosis cases and climatic variables from the Adelaide metropolitan area over the period 1990-2003, four regression methods were examined: standard Poisson regression, autoregressive adjusted Poisson regression, multiple linear regression, and a seasonal autoregressive integrated moving average (SARIMA) model. Notified salmonellosis cases in 2004 were used to test the forecasting ability of the four models. Parameter estimation, goodness-of-fit and forecasting ability of the four regression models were compared. Temperatures occurring 2 weeks prior to cases were positively associated with cases of salmonellosis. Rainfall was also inversely related to the number of cases. The comparison of the goodness-of-fit and forecasting ability suggest that the SARIMA model is better than the other three regression models. Temperature and rainfall may be used as climatic predictors of salmonellosis cases in regions with climatic characteristics similar to those of Adelaide. The SARIMA model could, thus, be adopted to quantify the relationship between climate variations and salmonellosis transmission.

SPSS macros to compare any two fitted values from a regression model.

PubMed

Weaver, Bruce; Dubois, Sacha

2012-12-01

In regression models with first-order terms only, the coefficient for a given variable is typically interpreted as the change in the fitted value of Y for a one-unit increase in that variable, with all other variables held constant. Therefore, each regression coefficient represents the difference between two fitted values of Y. But the coefficients represent only a fraction of the possible fitted value comparisons that might be of interest to researchers. For many fitted value comparisons that are not captured by any of the regression coefficients, common statistical software packages do not provide the standard errors needed to compute confidence intervals or carry out statistical tests-particularly in more complex models that include interactions, polynomial terms, or regression splines. We describe two SPSS macros that implement a matrix algebra method for comparing any two fitted values from a regression model. The !OLScomp and !MLEcomp macros are for use with models fitted via ordinary least squares and maximum likelihood estimation, respectively. The output from the macros includes the standard error of the difference between the two fitted values, a 95% confidence interval for the difference, and a corresponding statistical test with its p-value.

Analyzing Multilevel Data: Comparing Findings from Hierarchical Linear Modeling and Ordinary Least Squares Regression

ERIC Educational Resources Information Center

Rocconi, Louis M.

2013-01-01

This study examined the differing conclusions one may come to depending upon the type of analysis chosen, hierarchical linear modeling or ordinary least squares (OLS) regression. To illustrate this point, this study examined the influences of seniors' self-reported critical thinking abilities three ways: (1) an OLS regression with the student…

Two-Year versus One-Year Head Start Program Impact: Addressing Selection Bias by Comparing Regression Modeling with Propensity Score Analysis

ERIC Educational Resources Information Center

Leow, Christine; Wen, Xiaoli; Korfmacher, Jon

2015-01-01

This article compares regression modeling and propensity score analysis as different types of statistical techniques used in addressing selection bias when estimating the impact of two-year versus one-year Head Start on children's school readiness. The analyses were based on the national Head Start secondary dataset. After controlling for…

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.

Comparison of random regression models with Legendre polynomials and linear splines for production traits and somatic cell score of Canadian Holstein cows.

PubMed

Bohmanova, J; Miglior, F; Jamrozik, J; Misztal, I; Sullivan, P G

2008-09-01

A random regression model with both random and fixed regressions fitted by Legendre polynomials of order 4 was compared with 3 alternative models fitting linear splines with 4, 5, or 6 knots. The effects common for all models were a herd-test-date effect, fixed regressions on days in milk (DIM) nested within region-age-season of calving class, and random regressions for additive genetic and permanent environmental effects. Data were test-day milk, fat and protein yields, and SCS recorded from 5 to 365 DIM during the first 3 lactations of Canadian Holstein cows. A random sample of 50 herds consisting of 96,756 test-day records was generated to estimate variance components within a Bayesian framework via Gibbs sampling. Two sets of genetic evaluations were subsequently carried out to investigate performance of the 4 models. Models were compared by graphical inspection of variance functions, goodness of fit, error of prediction of breeding values, and stability of estimated breeding values. Models with splines gave lower estimates of variances at extremes of lactations than the model with Legendre polynomials. Differences among models in goodness of fit measured by percentages of squared bias, correlations between predicted and observed records, and residual variances were small. The deviance information criterion favored the spline model with 6 knots. Smaller error of prediction and higher stability of estimated breeding values were achieved by using spline models with 5 and 6 knots compared with the model with Legendre polynomials. In general, the spline model with 6 knots had the best overall performance based upon the considered model comparison criteria.

Modeling absolute differences in life expectancy with a censored skew-normal regression approach

PubMed Central

Clough-Gorr, Kerri; Zwahlen, Marcel

2015-01-01

Parameter estimates from commonly used multivariable parametric survival regression models do not directly quantify differences in years of life expectancy. Gaussian linear regression models give results in terms of absolute mean differences, but are not appropriate in modeling life expectancy, because in many situations time to death has a negative skewed distribution. A regression approach using a skew-normal distribution would be an alternative to parametric survival models in the modeling of life expectancy, because parameter estimates can be interpreted in terms of survival time differences while allowing for skewness of the distribution. In this paper we show how to use the skew-normal regression so that censored and left-truncated observations are accounted for. With this we model differences in life expectancy using data from the Swiss National Cohort Study and from official life expectancy estimates and compare the results with those derived from commonly used survival regression models. We conclude that a censored skew-normal survival regression approach for left-truncated observations can be used to model differences in life expectancy across covariates of interest. PMID:26339544

Estimation of variance in Cox's regression model with shared gamma frailties.

PubMed

Andersen, P K; Klein, J P; Knudsen, K M; Tabanera y Palacios, R

1997-12-01

The Cox regression model with a shared frailty factor allows for unobserved heterogeneity or for statistical dependence between the observed survival times. Estimation in this model when the frailties are assumed to follow a gamma distribution is reviewed, and we address the problem of obtaining variance estimates for regression coefficients, frailty parameter, and cumulative baseline hazards using the observed nonparametric information matrix. A number of examples are given comparing this approach with fully parametric inference in models with piecewise constant baseline hazards.

Comparison Between Linear and Non-parametric Regression Models for Genome-Enabled Prediction in Wheat

PubMed Central

Pérez-Rodríguez, Paulino; Gianola, Daniel; González-Camacho, Juan Manuel; Crossa, José; Manès, Yann; Dreisigacker, Susanne

2012-01-01

In genome-enabled prediction, parametric, semi-parametric, and non-parametric regression models have been used. This study assessed the predictive ability of linear and non-linear models using dense molecular markers. The linear models were linear on marker effects and included the Bayesian LASSO, Bayesian ridge regression, Bayes A, and Bayes B. The non-linear models (this refers to non-linearity on markers) were reproducing kernel Hilbert space (RKHS) regression, Bayesian regularized neural networks (BRNN), and radial basis function neural networks (RBFNN). These statistical models were compared using 306 elite wheat lines from CIMMYT genotyped with 1717 diversity array technology (DArT) markers and two traits, days to heading (DTH) and grain yield (GY), measured in each of 12 environments. It was found that the three non-linear models had better overall prediction accuracy than the linear regression specification. Results showed a consistent superiority of RKHS and RBFNN over the Bayesian LASSO, Bayesian ridge regression, Bayes A, and Bayes B models. PMID:23275882

Comparison between linear and non-parametric regression models for genome-enabled prediction in wheat.

PubMed

Pérez-Rodríguez, Paulino; Gianola, Daniel; González-Camacho, Juan Manuel; Crossa, José; Manès, Yann; Dreisigacker, Susanne

2012-12-01

In genome-enabled prediction, parametric, semi-parametric, and non-parametric regression models have been used. This study assessed the predictive ability of linear and non-linear models using dense molecular markers. The linear models were linear on marker effects and included the Bayesian LASSO, Bayesian ridge regression, Bayes A, and Bayes B. The non-linear models (this refers to non-linearity on markers) were reproducing kernel Hilbert space (RKHS) regression, Bayesian regularized neural networks (BRNN), and radial basis function neural networks (RBFNN). These statistical models were compared using 306 elite wheat lines from CIMMYT genotyped with 1717 diversity array technology (DArT) markers and two traits, days to heading (DTH) and grain yield (GY), measured in each of 12 environments. It was found that the three non-linear models had better overall prediction accuracy than the linear regression specification. Results showed a consistent superiority of RKHS and RBFNN over the Bayesian LASSO, Bayesian ridge regression, Bayes A, and Bayes B models.

Accounting for measurement error in log regression models with applications to accelerated testing.

PubMed

Richardson, Robert; Tolley, H Dennis; Evenson, William E; Lunt, Barry M

2018-01-01

In regression settings, parameter estimates will be biased when the explanatory variables are measured with error. This bias can significantly affect modeling goals. In particular, accelerated lifetime testing involves an extrapolation of the fitted model, and a small amount of bias in parameter estimates may result in a significant increase in the bias of the extrapolated predictions. Additionally, bias may arise when the stochastic component of a log regression model is assumed to be multiplicative when the actual underlying stochastic component is additive. To account for these possible sources of bias, a log regression model with measurement error and additive error is approximated by a weighted regression model which can be estimated using Iteratively Re-weighted Least Squares. Using the reduced Eyring equation in an accelerated testing setting, the model is compared to previously accepted approaches to modeling accelerated testing data with both simulations and real data.

Using ROC curves to compare neural networks and logistic regression for modeling individual noncatastrophic tree mortality

Treesearch

Susan L. King

2003-01-01

The performance of two classifiers, logistic regression and neural networks, are compared for modeling noncatastrophic individual tree mortality for 21 species of trees in West Virginia. The output of the classifier is usually a continuous number between 0 and 1. A threshold is selected between 0 and 1 and all of the trees below the threshold are classified as...

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

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.

Modelling of capital asset pricing by considering the lagged effects

NASA Astrophysics Data System (ADS)

Sukono; Hidayat, Y.; Bon, A. Talib bin; Supian, S.

2017-01-01

In this paper the problem of modelling the Capital Asset Pricing Model (CAPM) with the effect of the lagged is discussed. It is assumed that asset returns are analysed influenced by the market return and the return of risk-free assets. To analyse the relationship between asset returns, the market return, and the return of risk-free assets, it is conducted by using a regression equation of CAPM, and regression equation of lagged distributed CAPM. Associated with the regression equation lagged CAPM distributed, this paper also developed a regression equation of Koyck transformation CAPM. Results of development show that the regression equation of Koyck transformation CAPM has advantages, namely simple as it only requires three parameters, compared with regression equation of lagged distributed CAPM.

Data mining of tree-based models to analyze freeway accident frequency.

PubMed

Chang, Li-Yen; Chen, Wen-Chieh

2005-01-01

Statistical models, such as Poisson or negative binomial regression models, have been employed to analyze vehicle accident frequency for many years. However, these models have their own model assumptions and pre-defined underlying relationship between dependent and independent variables. If these assumptions are violated, the model could lead to erroneous estimation of accident likelihood. Classification and Regression Tree (CART), one of the most widely applied data mining techniques, has been commonly employed in business administration, industry, and engineering. CART does not require any pre-defined underlying relationship between target (dependent) variable and predictors (independent variables) and has been shown to be a powerful tool, particularly for dealing with prediction and classification problems. This study collected the 2001-2002 accident data of National Freeway 1 in Taiwan. A CART model and a negative binomial regression model were developed to establish the empirical relationship between traffic accidents and highway geometric variables, traffic characteristics, and environmental factors. The CART findings indicated that the average daily traffic volume and precipitation variables were the key determinants for freeway accident frequencies. By comparing the prediction performance between the CART and the negative binomial regression models, this study demonstrates that CART is a good alternative method for analyzing freeway accident frequencies. By comparing the prediction performance between the CART and the negative binomial regression models, this study demonstrates that CART is a good alternative method for analyzing freeway accident frequencies.

Zero-inflated Conway-Maxwell Poisson Distribution to Analyze Discrete Data.

PubMed

Sim, Shin Zhu; Gupta, Ramesh C; Ong, Seng Huat

2018-01-09

In this paper, we study the zero-inflated Conway-Maxwell Poisson (ZICMP) distribution and develop a regression model. Score and likelihood ratio tests are also implemented for testing the inflation/deflation parameter. Simulation studies are carried out to examine the performance of these tests. A data example is presented to illustrate the concepts. In this example, the proposed model is compared to the well-known zero-inflated Poisson (ZIP) and the zero- inflated generalized Poisson (ZIGP) regression models. It is shown that the fit by ZICMP is comparable or better than these models.

Linear regression based on Minimum Covariance Determinant (MCD) and TELBS methods on the productivity of phytoplankton

NASA Astrophysics Data System (ADS)

Gusriani, N.; Firdaniza

2018-03-01

The existence of outliers on multiple linear regression analysis causes the Gaussian assumption to be unfulfilled. If the Least Square method is forcedly used on these data, it will produce a model that cannot represent most data. For that, we need a robust regression method against outliers. This paper will compare the Minimum Covariance Determinant (MCD) method and the TELBS method on secondary data on the productivity of phytoplankton, which contains outliers. Based on the robust determinant coefficient value, MCD method produces a better model compared to TELBS method.

Two levels ARIMAX and regression models for forecasting time series data with calendar variation effects

NASA Astrophysics Data System (ADS)

Suhartono, Lee, Muhammad Hisyam; Prastyo, Dedy Dwi

2015-12-01

The aim of this research is to develop a calendar variation model for forecasting retail sales data with the Eid ul-Fitr effect. The proposed model is based on two methods, namely two levels ARIMAX and regression methods. Two levels ARIMAX and regression models are built by using ARIMAX for the first level and regression for the second level. Monthly men's jeans and women's trousers sales in a retail company for the period January 2002 to September 2009 are used as case study. In general, two levels of calendar variation model yields two models, namely the first model to reconstruct the sales pattern that already occurred, and the second model to forecast the effect of increasing sales due to Eid ul-Fitr that affected sales at the same and the previous months. The results show that the proposed two level calendar variation model based on ARIMAX and regression methods yields better forecast compared to the seasonal ARIMA model and Neural Networks.

A novel strategy for forensic age prediction by DNA methylation and support vector regression model

PubMed Central

Xu, Cheng; Qu, Hongzhu; Wang, Guangyu; Xie, Bingbing; Shi, Yi; Yang, Yaran; Zhao, Zhao; Hu, Lan; Fang, Xiangdong; Yan, Jiangwei; Feng, Lei

2015-01-01

High deviations resulting from prediction model, gender and population difference have limited age estimation application of DNA methylation markers. Here we identified 2,957 novel age-associated DNA methylation sites (P < 0.01 and R2 > 0.5) in blood of eight pairs of Chinese Han female monozygotic twins. Among them, nine novel sites (false discovery rate < 0.01), along with three other reported sites, were further validated in 49 unrelated female volunteers with ages of 20–80 years by Sequenom Massarray. A total of 95 CpGs were covered in the PCR products and 11 of them were built the age prediction models. After comparing four different models including, multivariate linear regression, multivariate nonlinear regression, back propagation neural network and support vector regression, SVR was identified as the most robust model with the least mean absolute deviation from real chronological age (2.8 years) and an average accuracy of 4.7 years predicted by only six loci from the 11 loci, as well as an less cross-validated error compared with linear regression model. Our novel strategy provides an accurate measurement that is highly useful in estimating the individual age in forensic practice as well as in tracking the aging process in other related applications. PMID:26635134

Random regression analyses using B-splines functions to model growth from birth to adult age in Canchim cattle.

PubMed

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

2010-12-01

The objective of this work was to estimate covariance functions using random regression models on B-splines functions of animal age, for weights from birth to adult age in Canchim cattle. Data comprised 49,011 records on 2435 females. The model of analysis included fixed effects of contemporary groups, age of dam as quadratic covariable and the population mean trend taken into account by a cubic regression on orthogonal polynomials of animal age. Residual variances were modelled through a step function with four classes. The direct and maternal additive genetic effects, and animal and maternal permanent environmental effects were included as random effects in the model. A total of seventeen analyses, considering linear, quadratic and cubic B-splines functions and up to seven knots, were carried out. B-spline functions of the same order were considered for all random effects. Random regression models on B-splines functions were compared to a random regression model on Legendre polynomials and with a multitrait model. Results from different models of analyses were compared using the REML form of the Akaike Information criterion and Schwarz' Bayesian Information criterion. In addition, the variance components and genetic parameters estimated for each random regression model were also used as criteria to choose the most adequate model to describe the covariance structure of the data. 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 adequate to describe the covariance structure of the data. Random regression models using B-spline functions as base functions fitted the data better than Legendre polynomials, especially at mature ages, but higher number of parameters need to be estimated with B-splines functions. © 2010 Blackwell Verlag GmbH.

Development and evaluation of habitat models for herpetofauna and small mammals

Treesearch

William M. Block; Michael L. Morrison; Peter E. Scott

1998-01-01

We evaluated the ability of discriminant analysis (DA), logistic regression (LR), and multiple regression (MR) to describe habitat use by amphibians, reptiles, and small mammals found in California oak woodlands. We also compared models derived from pitfall and live trapping data for several species. Habitat relations modeled by DA and LR produced similar results,...

Using Evidence-Based Decision Trees Instead of Formulas to Identify At-Risk Readers. REL 2014-036

ERIC Educational Resources Information Center

Koon, Sharon; Petscher, Yaacov; Foorman, Barbara R.

2014-01-01

This study examines whether the classification and regression tree (CART) model improves the early identification of students at risk for reading comprehension difficulties compared with the more difficult to interpret logistic regression model. CART is a type of predictive modeling that relies on nonparametric techniques. It presents results in…

Comparison and validation of statistical methods for predicting power outage durations in the event of hurricanes.

PubMed

Nateghi, Roshanak; Guikema, Seth D; Quiring, Steven M

2011-12-01

This article compares statistical methods for modeling power outage durations during hurricanes and examines the predictive accuracy of these methods. Being able to make accurate predictions of power outage durations is valuable because the information can be used by utility companies to plan their restoration efforts more efficiently. This information can also help inform customers and public agencies of the expected outage times, enabling better collective response planning, and coordination of restoration efforts for other critical infrastructures that depend on electricity. In the long run, outage duration estimates for future storm scenarios may help utilities and public agencies better allocate risk management resources to balance the disruption from hurricanes with the cost of hardening power systems. We compare the out-of-sample predictive accuracy of five distinct statistical models for estimating power outage duration times caused by Hurricane Ivan in 2004. The methods compared include both regression models (accelerated failure time (AFT) and Cox proportional hazard models (Cox PH)) and data mining techniques (regression trees, Bayesian additive regression trees (BART), and multivariate additive regression splines). We then validate our models against two other hurricanes. Our results indicate that BART yields the best prediction accuracy and that it is possible to predict outage durations with reasonable accuracy. © 2011 Society for Risk Analysis.

Modelling daily water temperature from air temperature for the Missouri River.

PubMed

Zhu, Senlin; Nyarko, Emmanuel Karlo; Hadzima-Nyarko, Marijana

2018-01-01

The bio-chemical and physical characteristics of a river are directly affected by water temperature, which thereby affects the overall health of aquatic ecosystems. It is a complex problem to accurately estimate water temperature. Modelling of river water temperature is usually based on a suitable mathematical model and field measurements of various atmospheric factors. In this article, the air-water temperature relationship of the Missouri River is investigated by developing three different machine learning models (Artificial Neural Network (ANN), Gaussian Process Regression (GPR), and Bootstrap Aggregated Decision Trees (BA-DT)). Standard models (linear regression, non-linear regression, and stochastic models) are also developed and compared to machine learning models. Analyzing the three standard models, the stochastic model clearly outperforms the standard linear model and nonlinear model. All the three machine learning models have comparable results and outperform the stochastic model, with GPR having slightly better results for stations No. 2 and 3, while BA-DT has slightly better results for station No. 1. The machine learning models are very effective tools which can be used for the prediction of daily river temperature.

Premium analysis for copula model: A case study for Malaysian motor insurance claims

NASA Astrophysics Data System (ADS)

Resti, Yulia; Ismail, Noriszura; Jaaman, Saiful Hafizah

2014-06-01

This study performs premium analysis for copula models with regression marginals. For illustration purpose, the copula models are fitted to the Malaysian motor insurance claims data. In this study, we consider copula models from Archimedean and Elliptical families, and marginal distributions of Gamma and Inverse Gaussian regression models. The simulated results from independent model, which is obtained from fitting regression models separately to each claim category, and dependent model, which is obtained from fitting copula models to all claim categories, are compared. The results show that the dependent model using Frank copula is the best model since the risk premiums estimated under this model are closely approximate to the actual claims experience relative to the other copula models.

Linear regression metamodeling as a tool to summarize and present simulation model results.

PubMed

Jalal, Hawre; Dowd, Bryan; Sainfort, François; Kuntz, Karen M

2013-10-01

Modelers lack a tool to systematically and clearly present complex model results, including those from sensitivity analyses. The objective was to propose linear regression metamodeling as a tool to increase transparency of decision analytic models and better communicate their results. We used a simplified cancer cure model to demonstrate our approach. The model computed the lifetime cost and benefit of 3 treatment options for cancer patients. We simulated 10,000 cohorts in a probabilistic sensitivity analysis (PSA) and regressed the model outcomes on the standardized input parameter values in a set of regression analyses. We used the regression coefficients to describe measures of sensitivity analyses, including threshold and parameter sensitivity analyses. We also compared the results of the PSA to deterministic full-factorial and one-factor-at-a-time designs. The regression intercept represented the estimated base-case outcome, and the other coefficients described the relative parameter uncertainty in the model. We defined simple relationships that compute the average and incremental net benefit of each intervention. Metamodeling produced outputs similar to traditional deterministic 1-way or 2-way sensitivity analyses but was more reliable since it used all parameter values. Linear regression metamodeling is a simple, yet powerful, tool that can assist modelers in communicating model characteristics and sensitivity analyses.

Retargeted Least Squares Regression Algorithm.

PubMed

Zhang, Xu-Yao; Wang, Lingfeng; Xiang, Shiming; Liu, Cheng-Lin

2015-09-01

This brief presents a framework of retargeted least squares regression (ReLSR) for multicategory classification. The core idea is to directly learn the regression targets from data other than using the traditional zero-one matrix as regression targets. The learned target matrix can guarantee a large margin constraint for the requirement of correct classification for each data point. Compared with the traditional least squares regression (LSR) and a recently proposed discriminative LSR models, ReLSR is much more accurate in measuring the classification error of the regression model. Furthermore, ReLSR is a single and compact model, hence there is no need to train two-class (binary) machines that are independent of each other. The convex optimization problem of ReLSR is solved elegantly and efficiently with an alternating procedure including regression and retargeting as substeps. The experimental evaluation over a range of databases identifies the validity of our method.

Antibiotic Resistances in Livestock: A Comparative Approach to Identify an Appropriate Regression Model for Count Data

PubMed Central

Hüls, Anke; Frömke, Cornelia; Ickstadt, Katja; Hille, Katja; Hering, Johanna; von Münchhausen, Christiane; Hartmann, Maria; Kreienbrock, Lothar

2017-01-01

Antimicrobial resistance in livestock is a matter of general concern. To develop hygiene measures and methods for resistance prevention and control, epidemiological studies on a population level are needed to detect factors associated with antimicrobial resistance in livestock holdings. In general, regression models are used to describe these relationships between environmental factors and resistance outcome. Besides the study design, the correlation structures of the different outcomes of antibiotic resistance and structural zero measurements on the resistance outcome as well as on the exposure side are challenges for the epidemiological model building process. The use of appropriate regression models that acknowledge these complexities is essential to assure valid epidemiological interpretations. The aims of this paper are (i) to explain the model building process comparing several competing models for count data (negative binomial model, quasi-Poisson model, zero-inflated model, and hurdle model) and (ii) to compare these models using data from a cross-sectional study on antibiotic resistance in animal husbandry. These goals are essential to evaluate which model is most suitable to identify potential prevention measures. The dataset used as an example in our analyses was generated initially to study the prevalence and associated factors for the appearance of cefotaxime-resistant Escherichia coli in 48 German fattening pig farms. For each farm, the outcome was the count of samples with resistant bacteria. There was almost no overdispersion and only moderate evidence of excess zeros in the data. Our analyses show that it is essential to evaluate regression models in studies analyzing the relationship between environmental factors and antibiotic resistances in livestock. After model comparison based on evaluation of model predictions, Akaike information criterion, and Pearson residuals, here the hurdle model was judged to be the most appropriate model. PMID:28620609

Evaluation of Regression Models of Balance Calibration Data Using an Empirical Criterion

NASA Technical Reports Server (NTRS)

Ulbrich, Norbert; Volden, Thomas R.

2012-01-01

An empirical criterion for assessing the significance of individual terms of regression models of wind tunnel strain gage balance outputs is evaluated. The criterion is based on the percent contribution of a regression model term. It considers a term to be significant if its percent contribution exceeds the empirical threshold of 0.05%. The criterion has the advantage that it can easily be computed using the regression coefficients of the gage outputs and the load capacities of the balance. First, a definition of the empirical criterion is provided. Then, it is compared with an alternate statistical criterion that is widely used in regression analysis. Finally, calibration data sets from a variety of balances are used to illustrate the connection between the empirical and the statistical criterion. A review of these results indicated that the empirical criterion seems to be suitable for a crude assessment of the significance of a regression model term as the boundary between a significant and an insignificant term cannot be defined very well. Therefore, regression model term reduction should only be performed by using the more universally applicable statistical criterion.

Impact of multicollinearity on small sample hydrologic regression models

NASA Astrophysics Data System (ADS)

Kroll, Charles N.; Song, Peter

2013-06-01

Often hydrologic regression models are developed with ordinary least squares (OLS) procedures. The use of OLS with highly correlated explanatory variables produces multicollinearity, which creates highly sensitive parameter estimators with inflated variances and improper model selection. It is not clear how to best address multicollinearity in hydrologic regression models. Here a Monte Carlo simulation is developed to compare four techniques to address multicollinearity: OLS, OLS with variance inflation factor screening (VIF), principal component regression (PCR), and partial least squares regression (PLS). The performance of these four techniques was observed for varying sample sizes, correlation coefficients between the explanatory variables, and model error variances consistent with hydrologic regional regression models. The negative effects of multicollinearity are magnified at smaller sample sizes, higher correlations between the variables, and larger model error variances (smaller R2). The Monte Carlo simulation indicates that if the true model is known, multicollinearity is present, and the estimation and statistical testing of regression parameters are of interest, then PCR or PLS should be employed. If the model is unknown, or if the interest is solely on model predictions, is it recommended that OLS be employed since using more complicated techniques did not produce any improvement in model performance. A leave-one-out cross-validation case study was also performed using low-streamflow data sets from the eastern United States. Results indicate that OLS with stepwise selection generally produces models across study regions with varying levels of multicollinearity that are as good as biased regression techniques such as PCR and PLS.

A comparison of time dependent Cox regression, pooled logistic regression and cross sectional pooling with simulations and an application to the Framingham Heart Study.

PubMed

Ngwa, Julius S; Cabral, Howard J; Cheng, Debbie M; Pencina, Michael J; Gagnon, David R; LaValley, Michael P; Cupples, L Adrienne

2016-11-03

Typical survival studies follow individuals to an event and measure explanatory variables for that event, sometimes repeatedly over the course of follow up. The Cox regression model has been used widely in the analyses of time to diagnosis or death from disease. The associations between the survival outcome and time dependent measures may be biased unless they are modeled appropriately. In this paper we explore the Time Dependent Cox Regression Model (TDCM), which quantifies the effect of repeated measures of covariates in the analysis of time to event data. This model is commonly used in biomedical research but sometimes does not explicitly adjust for the times at which time dependent explanatory variables are measured. This approach can yield different estimates of association compared to a model that adjusts for these times. In order to address the question of how different these estimates are from a statistical perspective, we compare the TDCM to Pooled Logistic Regression (PLR) and Cross Sectional Pooling (CSP), considering models that adjust and do not adjust for time in PLR and CSP. In a series of simulations we found that time adjusted CSP provided identical results to the TDCM while the PLR showed larger parameter estimates compared to the time adjusted CSP and the TDCM in scenarios with high event rates. We also observed upwardly biased estimates in the unadjusted CSP and unadjusted PLR methods. The time adjusted PLR had a positive bias in the time dependent Age effect with reduced bias when the event rate is low. The PLR methods showed a negative bias in the Sex effect, a subject level covariate, when compared to the other methods. The Cox models yielded reliable estimates for the Sex effect in all scenarios considered. We conclude that survival analyses that explicitly account in the statistical model for the times at which time dependent covariates are measured provide more reliable estimates compared to unadjusted analyses. We present results from the Framingham Heart Study in which lipid measurements and myocardial infarction data events were collected over a period of 26 years.

Estimation and Testing of Partial Covariances, Correlations, and Regression Weights Using Maximum Likelihood Factor Analysis.

ERIC Educational Resources Information Center

And Others; Werts, Charles E.

1979-01-01

It is shown how partial covariance, part and partial correlation, and regression weights can be estimated and tested for significance by means of a factor analytic model. Comparable partial covariance, correlations, and regression weights have identical significance tests. (Author)

RBF kernel based support vector regression to estimate the blood volume and heart rate responses during hemodialysis.

PubMed

Javed, Faizan; Chan, Gregory S H; Savkin, Andrey V; Middleton, Paul M; Malouf, Philip; Steel, Elizabeth; Mackie, James; Lovell, Nigel H

2009-01-01

This paper uses non-linear support vector regression (SVR) to model the blood volume and heart rate (HR) responses in 9 hemodynamically stable kidney failure patients during hemodialysis. Using radial bias function (RBF) kernels the non-parametric models of relative blood volume (RBV) change with time as well as percentage change in HR with respect to RBV were obtained. The e-insensitivity based loss function was used for SVR modeling. Selection of the design parameters which includes capacity (C), insensitivity region (e) and the RBF kernel parameter (sigma) was made based on a grid search approach and the selected models were cross-validated using the average mean square error (AMSE) calculated from testing data based on a k-fold cross-validation technique. Linear regression was also applied to fit the curves and the AMSE was calculated for comparison with SVR. For the model based on RBV with time, SVR gave a lower AMSE for both training (AMSE=1.5) as well as testing data (AMSE=1.4) compared to linear regression (AMSE=1.8 and 1.5). SVR also provided a better fit for HR with RBV for both training as well as testing data (AMSE=15.8 and 16.4) compared to linear regression (AMSE=25.2 and 20.1).

Comparing The Effectiveness of a90/95 Calculations (Preprint)

DTIC Science & Technology

2006-09-01

Nachtsheim, John Neter, William Li, Applied Linear Statistical Models , 5th ed., McGraw-Hill/Irwin, 2005 5. Mood, Graybill and Boes, Introduction...curves is based on methods that are only valid for ordinary linear regression. Requirements for a valid Ordinary Least-Squares Regression Model There... linear . For example is a linear model ; is not. 2. Uniform variance (homoscedasticity

A Comparative Investigation of the Combined Effects of Pre-Processing, Wavelength Selection, and Regression Methods on Near-Infrared Calibration Model Performance.

PubMed

Wan, Jian; Chen, Yi-Chieh; Morris, A Julian; Thennadil, Suresh N

2017-07-01

Near-infrared (NIR) spectroscopy is being widely used in various fields ranging from pharmaceutics to the food industry for analyzing chemical and physical properties of the substances concerned. Its advantages over other analytical techniques include available physical interpretation of spectral data, nondestructive nature and high speed of measurements, and little or no need for sample preparation. The successful application of NIR spectroscopy relies on three main aspects: pre-processing of spectral data to eliminate nonlinear variations due to temperature, light scattering effects and many others, selection of those wavelengths that contribute useful information, and identification of suitable calibration models using linear/nonlinear regression . Several methods have been developed for each of these three aspects and many comparative studies of different methods exist for an individual aspect or some combinations. However, there is still a lack of comparative studies for the interactions among these three aspects, which can shed light on what role each aspect plays in the calibration and how to combine various methods of each aspect together to obtain the best calibration model. This paper aims to provide such a comparative study based on four benchmark data sets using three typical pre-processing methods, namely, orthogonal signal correction (OSC), extended multiplicative signal correction (EMSC) and optical path-length estimation and correction (OPLEC); two existing wavelength selection methods, namely, stepwise forward selection (SFS) and genetic algorithm optimization combined with partial least squares regression for spectral data (GAPLSSP); four popular regression methods, namely, partial least squares (PLS), least absolute shrinkage and selection operator (LASSO), least squares support vector machine (LS-SVM), and Gaussian process regression (GPR). The comparative study indicates that, in general, pre-processing of spectral data can play a significant role in the calibration while wavelength selection plays a marginal role and the combination of certain pre-processing, wavelength selection, and nonlinear regression methods can achieve superior performance over traditional linear regression-based calibration.

Comparing Methodologies for Developing an Early Warning System: Classification and Regression Tree Model versus Logistic Regression. REL 2015-077

ERIC Educational Resources Information Center

Koon, Sharon; Petscher, Yaacov

2015-01-01

The purpose of this report was to explicate the use of logistic regression and classification and regression tree (CART) analysis in the development of early warning systems. It was motivated by state education leaders' interest in maintaining high classification accuracy while simultaneously improving practitioner understanding of the rules by…

Evaluation of methodology for the analysis of 'time-to-event' data in pharmacogenomic genome-wide association studies.

PubMed

Syed, Hamzah; Jorgensen, Andrea L; Morris, Andrew P

2016-06-01

To evaluate the power to detect associations between SNPs and time-to-event outcomes across a range of pharmacogenomic study designs while comparing alternative regression approaches. Simulations were conducted to compare Cox proportional hazards modeling accounting for censoring and logistic regression modeling of a dichotomized outcome at the end of the study. The Cox proportional hazards model was demonstrated to be more powerful than the logistic regression analysis. The difference in power between the approaches was highly dependent on the rate of censoring. Initial evaluation of single-nucleotide polymorphism association signals using computationally efficient software with dichotomized outcomes provides an effective screening tool for some design scenarios, and thus has important implications for the development of analytical protocols in pharmacogenomic studies.

Wavelet regression model in forecasting crude oil price

NASA Astrophysics Data System (ADS)

Hamid, Mohd Helmie; Shabri, Ani

2017-05-01

This study presents the performance of wavelet multiple linear regression (WMLR) technique in daily crude oil forecasting. WMLR model was developed by integrating the discrete wavelet transform (DWT) and multiple linear regression (MLR) model. The original time series was decomposed to sub-time series with different scales by wavelet theory. Correlation analysis was conducted to assist in the selection of optimal decomposed components as inputs for the WMLR model. The daily WTI crude oil price series has been used in this study to test the prediction capability of the proposed model. The forecasting performance of WMLR model were also compared with regular multiple linear regression (MLR), Autoregressive Moving Average (ARIMA) and Generalized Autoregressive Conditional Heteroscedasticity (GARCH) using root mean square errors (RMSE) and mean absolute errors (MAE). Based on the experimental results, it appears that the WMLR model performs better than the other forecasting technique tested in this study.

A Comparison of Conventional Linear Regression Methods and Neural Networks for Forecasting Educational Spending.

ERIC Educational Resources Information Center

Baker, Bruce D.; Richards, Craig E.

1999-01-01

Applies neural network methods for forecasting 1991-95 per-pupil expenditures in U.S. public elementary and secondary schools. Forecasting models included the National Center for Education Statistics' multivariate regression model and three neural architectures. Regarding prediction accuracy, neural network results were comparable or superior to…

The Development and Demonstration of Multiple Regression Models for Operant Conditioning Questions.

ERIC Educational Resources Information Center

Fanning, Fred; Newman, Isadore

Based on the assumption that inferential statistics can make the operant conditioner more sensitive to possible significant relationships, regressions models were developed to test the statistical significance between slopes and Y intercepts of the experimental and control group subjects. These results were then compared to the traditional operant…

A comparative study between nonlinear regression and artificial neural network approaches for modelling wild oat (Avena fatua) field emergence

USDA-ARS?s Scientific Manuscript database

Non-linear regression techniques are used widely to fit weed field emergence patterns to soil microclimatic indices using S-type functions. Artificial neural networks present interesting and alternative features for such modeling purposes. In this work, a univariate hydrothermal-time based Weibull m...

A Comparison of Methods for Estimating Quadratic Effects in Nonlinear Structural Equation Models

ERIC Educational Resources Information Center

Harring, Jeffrey R.; Weiss, Brandi A.; Hsu, Jui-Chen

2012-01-01

Two Monte Carlo simulations were performed to compare methods for estimating and testing hypotheses of quadratic effects in latent variable regression models. The methods considered in the current study were (a) a 2-stage moderated regression approach using latent variable scores, (b) an unconstrained product indicator approach, (c) a latent…

OPLS statistical model versus linear regression to assess sonographic predictors of stroke prognosis.

PubMed

Vajargah, Kianoush Fathi; Sadeghi-Bazargani, Homayoun; Mehdizadeh-Esfanjani, Robab; Savadi-Oskouei, Daryoush; Farhoudi, Mehdi

2012-01-01

The objective of the present study was to assess the comparable applicability of orthogonal projections to latent structures (OPLS) statistical model vs traditional linear regression in order to investigate the role of trans cranial doppler (TCD) sonography in predicting ischemic stroke prognosis. The study was conducted on 116 ischemic stroke patients admitted to a specialty neurology ward. The Unified Neurological Stroke Scale was used once for clinical evaluation on the first week of admission and again six months later. All data was primarily analyzed using simple linear regression and later considered for multivariate analysis using PLS/OPLS models through the SIMCA P+12 statistical software package. The linear regression analysis results used for the identification of TCD predictors of stroke prognosis were confirmed through the OPLS modeling technique. Moreover, in comparison to linear regression, the OPLS model appeared to have higher sensitivity in detecting the predictors of ischemic stroke prognosis and detected several more predictors. Applying the OPLS model made it possible to use both single TCD measures/indicators and arbitrarily dichotomized measures of TCD single vessel involvement as well as the overall TCD result. In conclusion, the authors recommend PLS/OPLS methods as complementary rather than alternative to the available classical regression models such as linear regression.

Completing the Remedial Sequence and College-Level Credit-Bearing Math: Comparing Binary, Cumulative, and Continuation Ratio Logistic Regression Models

ERIC Educational Resources Information Center

Davidson, J. Cody

2016-01-01

Mathematics is the most common subject area of remedial need and the majority of remedial math students never pass a college-level credit-bearing math class. The majorities of studies that investigate this phenomenon are conducted at community colleges and use some type of regression model; however, none have used a continuation ratio model. The…

Regression modeling and mapping of coniferous forest basal area and tree density from discrete-return lidar and multispectral data

Treesearch

Andrew T. Hudak; Nicholas L. Crookston; Jeffrey S. Evans; Michael K. Falkowski; Alistair M. S. Smith; Paul E. Gessler; Penelope Morgan

2006-01-01

We compared the utility of discrete-return light detection and ranging (lidar) data and multispectral satellite imagery, and their integration, for modeling and mapping basal area and tree density across two diverse coniferous forest landscapes in north-central Idaho. We applied multiple linear regression models subset from a suite of 26 predictor variables derived...

[Comparison of predictive effect between the single auto regressive integrated moving average (ARIMA) model and the ARIMA-generalized regression neural network (GRNN) combination model on the incidence of scarlet fever].

PubMed

Zhu, Yu; Xia, Jie-lai; Wang, Jing

2009-09-01

Application of the 'single auto regressive integrated moving average (ARIMA) model' and the 'ARIMA-generalized regression neural network (GRNN) combination model' in the research of the incidence of scarlet fever. Establish the auto regressive integrated moving average model based on the data of the monthly incidence on scarlet fever of one city, from 2000 to 2006. The fitting values of the ARIMA model was used as input of the GRNN, and the actual values were used as output of the GRNN. After training the GRNN, the effect of the single ARIMA model and the ARIMA-GRNN combination model was then compared. The mean error rate (MER) of the single ARIMA model and the ARIMA-GRNN combination model were 31.6%, 28.7% respectively and the determination coefficient (R(2)) of the two models were 0.801, 0.872 respectively. The fitting efficacy of the ARIMA-GRNN combination model was better than the single ARIMA, which had practical value in the research on time series data such as the incidence of scarlet fever.

Data Mining Methods Applied to Flight Operations Quality Assurance Data: A Comparison to Standard Statistical Methods

NASA Technical Reports Server (NTRS)

Stolzer, Alan J.; Halford, Carl

2007-01-01

In a previous study, multiple regression techniques were applied to Flight Operations Quality Assurance-derived data to develop parsimonious model(s) for fuel consumption on the Boeing 757 airplane. The present study examined several data mining algorithms, including neural networks, on the fuel consumption problem and compared them to the multiple regression results obtained earlier. Using regression methods, parsimonious models were obtained that explained approximately 85% of the variation in fuel flow. In general data mining methods were more effective in predicting fuel consumption. Classification and Regression Tree methods reported correlation coefficients of .91 to .92, and General Linear Models and Multilayer Perceptron neural networks reported correlation coefficients of about .99. These data mining models show great promise for use in further examining large FOQA databases for operational and safety improvements.

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.

Robust analysis of trends in noisy tokamak confinement data using geodesic least squares regression

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

Verdoolaege, G., E-mail: geert.verdoolaege@ugent.be; Laboratory for Plasma Physics, Royal Military Academy, B-1000 Brussels; Shabbir, A.

Regression analysis is a very common activity in fusion science for unveiling trends and parametric dependencies, but it can be a difficult matter. We have recently developed the method of geodesic least squares (GLS) regression that is able to handle errors in all variables, is robust against data outliers and uncertainty in the regression model, and can be used with arbitrary distribution models and regression functions. We here report on first results of application of GLS to estimation of the multi-machine scaling law for the energy confinement time in tokamaks, demonstrating improved consistency of the GLS results compared to standardmore » least squares.« less

Predicting Final GPA of Graduate School Students: Comparing Artificial Neural Networking and Simultaneous Multiple Regression

ERIC Educational Resources Information Center

Anderson, Joan L.

2006-01-01

Data from graduate student applications at a large Western university were used to determine which factors were the best predictors of success in graduate school, as defined by cumulative graduate grade point average. Two statistical models were employed and compared: artificial neural networking and simultaneous multiple regression. Both models…

Regression Model Optimization for the Analysis of Experimental Data

NASA Technical Reports Server (NTRS)

Ulbrich, N.

2009-01-01

A candidate math model search algorithm was developed at Ames Research Center that determines a recommended math model for the multivariate regression analysis of experimental data. The search algorithm is applicable to classical regression analysis problems as well as wind tunnel strain gage balance calibration analysis applications. The algorithm compares the predictive capability of different regression models using the standard deviation of the PRESS residuals of the responses as a search metric. This search metric is minimized during the search. Singular value decomposition is used during the search to reject math models that lead to a singular solution of the regression analysis problem. Two threshold dependent constraints are also applied. The first constraint rejects math models with insignificant terms. The second constraint rejects math models with near-linear dependencies between terms. The math term hierarchy rule may also be applied as an optional constraint during or after the candidate math model search. The final term selection of the recommended math model depends on the regressor and response values of the data set, the user s function class combination choice, the user s constraint selections, and the result of the search metric minimization. A frequently used regression analysis example from the literature is used to illustrate the application of the search algorithm to experimental data.

Pseudo-second order models for the adsorption of safranin onto activated carbon: comparison of linear and non-linear regression methods.

PubMed

Kumar, K Vasanth

2007-04-02

Kinetic experiments were carried out for the sorption of safranin onto activated carbon particles. The kinetic data were fitted to pseudo-second order model of Ho, Sobkowsk and Czerwinski, Blanchard et al. and Ritchie by linear and non-linear regression methods. Non-linear method was found to be a better way of obtaining the parameters involved in the second order rate kinetic expressions. Both linear and non-linear regression showed that the Sobkowsk and Czerwinski and Ritchie's pseudo-second order models were the same. Non-linear regression analysis showed that both Blanchard et al. and Ho have similar ideas on the pseudo-second order model but with different assumptions. The best fit of experimental data in Ho's pseudo-second order expression by linear and non-linear regression method showed that Ho pseudo-second order model was a better kinetic expression when compared to other pseudo-second order kinetic expressions.

Genetic Programming Transforms in Linear Regression Situations

NASA Astrophysics Data System (ADS)

Castillo, Flor; Kordon, Arthur; Villa, Carlos

The chapter summarizes the use of Genetic Programming (GP) inMultiple Linear Regression (MLR) to address multicollinearity and Lack of Fit (LOF). The basis of the proposed method is applying appropriate input transforms (model respecification) that deal with these issues while preserving the information content of the original variables. The transforms are selected from symbolic regression models with optimal trade-off between accuracy of prediction and expressional complexity, generated by multiobjective Pareto-front GP. The chapter includes a comparative study of the GP-generated transforms with Ridge Regression, a variant of ordinary Multiple Linear Regression, which has been a useful and commonly employed approach for reducing multicollinearity. The advantages of GP-generated model respecification are clearly defined and demonstrated. Some recommendations for transforms selection are given as well. The application benefits of the proposed approach are illustrated with a real industrial application in one of the broadest empirical modeling areas in manufacturing - robust inferential sensors. The chapter contributes to increasing the awareness of the potential of GP in statistical model building by MLR.

Partial Least Squares with Structured Output for Modelling the Metabolomics Data Obtained from Complex Experimental Designs: A Study into the Y-Block Coding.

PubMed

Xu, Yun; Muhamadali, Howbeer; Sayqal, Ali; Dixon, Neil; Goodacre, Royston

2016-10-28

Partial least squares (PLS) is one of the most commonly used supervised modelling approaches for analysing multivariate metabolomics data. PLS is typically employed as either a regression model (PLS-R) or a classification model (PLS-DA). However, in metabolomics studies it is common to investigate multiple, potentially interacting, factors simultaneously following a specific experimental design. Such data often cannot be considered as a "pure" regression or a classification problem. Nevertheless, these data have often still been treated as a regression or classification problem and this could lead to ambiguous results. In this study, we investigated the feasibility of designing a hybrid target matrix Y that better reflects the experimental design than simple regression or binary class membership coding commonly used in PLS modelling. The new design of Y coding was based on the same principle used by structural modelling in machine learning techniques. Two real metabolomics datasets were used as examples to illustrate how the new Y coding can improve the interpretability of the PLS model compared to classic regression/classification coding.

Comparison and Contrast of Two General Functional Regression Modeling Frameworks

PubMed Central

Morris, Jeffrey S.

2017-01-01

In this article, Greven and Scheipl describe an impressively general framework for performing functional regression that builds upon the generalized additive modeling framework. Over the past number of years, my collaborators and I have also been developing a general framework for functional regression, functional mixed models, which shares many similarities with this framework, but has many differences as well. In this discussion, I compare and contrast these two frameworks, to hopefully illuminate characteristics of each, highlighting their respecitve strengths and weaknesses, and providing recommendations regarding the settings in which each approach might be preferable. PMID:28736502

Comparison and Contrast of Two General Functional Regression Modeling Frameworks.

PubMed

Morris, Jeffrey S

2017-02-01

In this article, Greven and Scheipl describe an impressively general framework for performing functional regression that builds upon the generalized additive modeling framework. Over the past number of years, my collaborators and I have also been developing a general framework for functional regression, functional mixed models, which shares many similarities with this framework, but has many differences as well. In this discussion, I compare and contrast these two frameworks, to hopefully illuminate characteristics of each, highlighting their respecitve strengths and weaknesses, and providing recommendations regarding the settings in which each approach might be preferable.

Detecting DIF in Polytomous Items Using MACS, IRT and Ordinal Logistic Regression

ERIC Educational Resources Information Center

Elosua, Paula; Wells, Craig

2013-01-01

The purpose of the present study was to compare the Type I error rate and power of two model-based procedures, the mean and covariance structure model (MACS) and the item response theory (IRT), and an observed-score based procedure, ordinal logistic regression, for detecting differential item functioning (DIF) in polytomous items. A simulation…

Anticipating Mathematics Performance: A Cross-Validation Comparison of AID3 and Regression. AIR 1988 Annual Forum Paper.

ERIC Educational Resources Information Center

Bloom, Allan M.; And Others

In response to the increasing importance of student performance in required classes, research was conducted to compare two prediction procedures, linear modeling using multiple regression and nonlinear modeling using AID3. Performance in the first college math course (College Mathematics, Calculus, or Business Calculus Matrices) was the dependent…

Forecasting daily patient volumes in the emergency department.

PubMed

Jones, Spencer S; Thomas, Alun; Evans, R Scott; Welch, Shari J; Haug, Peter J; Snow, Gregory L

2008-02-01

Shifts in the supply of and demand for emergency department (ED) resources make the efficient allocation of ED resources increasingly important. Forecasting is a vital activity that guides decision-making in many areas of economic, industrial, and scientific planning, but has gained little traction in the health care industry. There are few studies that explore the use of forecasting methods to predict patient volumes in the ED. The goals of this study are to explore and evaluate the use of several statistical forecasting methods to predict daily ED patient volumes at three diverse hospital EDs and to compare the accuracy of these methods to the accuracy of a previously proposed forecasting method. Daily patient arrivals at three hospital EDs were collected for the period January 1, 2005, through March 31, 2007. The authors evaluated the use of seasonal autoregressive integrated moving average, time series regression, exponential smoothing, and artificial neural network models to forecast daily patient volumes at each facility. Forecasts were made for horizons ranging from 1 to 30 days in advance. The forecast accuracy achieved by the various forecasting methods was compared to the forecast accuracy achieved when using a benchmark forecasting method already available in the emergency medicine literature. All time series methods considered in this analysis provided improved in-sample model goodness of fit. However, post-sample analysis revealed that time series regression models that augment linear regression models by accounting for serial autocorrelation offered only small improvements in terms of post-sample forecast accuracy, relative to multiple linear regression models, while seasonal autoregressive integrated moving average, exponential smoothing, and artificial neural network forecasting models did not provide consistently accurate forecasts of daily ED volumes. This study confirms the widely held belief that daily demand for ED services is characterized by seasonal and weekly patterns. The authors compared several time series forecasting methods to a benchmark multiple linear regression model. The results suggest that the existing methodology proposed in the literature, multiple linear regression based on calendar variables, is a reasonable approach to forecasting daily patient volumes in the ED. However, the authors conclude that regression-based models that incorporate calendar variables, account for site-specific special-day effects, and allow for residual autocorrelation provide a more appropriate, informative, and consistently accurate approach to forecasting daily ED patient volumes.

Landslide susceptibility mapping using frequency ratio, logistic regression, artificial neural networks and their comparison: A case study from Kat landslides (Tokat—Turkey)

NASA Astrophysics Data System (ADS)

Yilmaz, Işık

2009-06-01

The purpose of this study is to compare the landslide susceptibility mapping methods of frequency ratio (FR), logistic regression and artificial neural networks (ANN) applied in the Kat County (Tokat—Turkey). Digital elevation model (DEM) was first constructed using GIS software. Landslide-related factors such as geology, faults, drainage system, topographical elevation, slope angle, slope aspect, topographic wetness index (TWI) and stream power index (SPI) were used in the landslide susceptibility analyses. Landslide susceptibility maps were produced from the frequency ratio, logistic regression and neural networks models, and they were then compared by means of their validations. The higher accuracies of the susceptibility maps for all three models were obtained from the comparison of the landslide susceptibility maps with the known landslide locations. However, respective area under curve (AUC) values of 0.826, 0.842 and 0.852 for frequency ratio, logistic regression and artificial neural networks showed that the map obtained from ANN model is more accurate than the other models, accuracies of all models can be evaluated relatively similar. The results obtained in this study also showed that the frequency ratio model can be used as a simple tool in assessment of landslide susceptibility when a sufficient number of data were obtained. Input process, calculations and output process are very simple and can be readily understood in the frequency ratio model, however logistic regression and neural networks require the conversion of data to ASCII or other formats. Moreover, it is also very hard to process the large amount of data in the statistical package.

Revisiting Gaussian Process Regression Modeling for Localization in Wireless Sensor Networks

PubMed Central

Richter, Philipp; Toledano-Ayala, Manuel

2015-01-01

Signal strength-based positioning in wireless sensor networks is a key technology for seamless, ubiquitous localization, especially in areas where Global Navigation Satellite System (GNSS) signals propagate poorly. To enable wireless local area network (WLAN) location fingerprinting in larger areas while maintaining accuracy, methods to reduce the effort of radio map creation must be consolidated and automatized. Gaussian process regression has been applied to overcome this issue, also with auspicious results, but the fit of the model was never thoroughly assessed. Instead, most studies trained a readily available model, relying on the zero mean and squared exponential covariance function, without further scrutinization. This paper studies the Gaussian process regression model selection for WLAN fingerprinting in indoor and outdoor environments. We train several models for indoor/outdoor- and combined areas; we evaluate them quantitatively and compare them by means of adequate model measures, hence assessing the fit of these models directly. To illuminate the quality of the model fit, the residuals of the proposed model are investigated, as well. Comparative experiments on the positioning performance verify and conclude the model selection. In this way, we show that the standard model is not the most appropriate, discuss alternatives and present our best candidate. PMID:26370996

Evaluation of logistic regression models and effect of covariates for case-control study in RNA-Seq analysis.

PubMed

Choi, Seung Hoan; Labadorf, Adam T; Myers, Richard H; Lunetta, Kathryn L; Dupuis, Josée; DeStefano, Anita L

2017-02-06

Next generation sequencing provides a count of RNA molecules in the form of short reads, yielding discrete, often highly non-normally distributed gene expression measurements. Although Negative Binomial (NB) regression has been generally accepted in the analysis of RNA sequencing (RNA-Seq) data, its appropriateness has not been exhaustively evaluated. We explore logistic regression as an alternative method for RNA-Seq studies designed to compare cases and controls, where disease status is modeled as a function of RNA-Seq reads using simulated and Huntington disease data. We evaluate the effect of adjusting for covariates that have an unknown relationship with gene expression. Finally, we incorporate the data adaptive method in order to compare false positive rates. When the sample size is small or the expression levels of a gene are highly dispersed, the NB regression shows inflated Type-I error rates but the Classical logistic and Bayes logistic (BL) regressions are conservative. Firth's logistic (FL) regression performs well or is slightly conservative. Large sample size and low dispersion generally make Type-I error rates of all methods close to nominal alpha levels of 0.05 and 0.01. However, Type-I error rates are controlled after applying the data adaptive method. The NB, BL, and FL regressions gain increased power with large sample size, large log2 fold-change, and low dispersion. The FL regression has comparable power to NB regression. We conclude that implementing the data adaptive method appropriately controls Type-I error rates in RNA-Seq analysis. Firth's logistic regression provides a concise statistical inference process and reduces spurious associations from inaccurately estimated dispersion parameters in the negative binomial framework.

Regression Models For Multivariate Count Data

PubMed Central

Zhang, Yiwen; Zhou, Hua; Zhou, Jin; Sun, Wei

2016-01-01

Data with multivariate count responses frequently occur in modern applications. The commonly used multinomial-logit model is limiting due to its restrictive mean-variance structure. For instance, analyzing count data from the recent RNA-seq technology by the multinomial-logit model leads to serious errors in hypothesis testing. The ubiquity of over-dispersion and complicated correlation structures among multivariate counts calls for more flexible regression models. In this article, we study some generalized linear models that incorporate various correlation structures among the counts. Current literature lacks a treatment of these models, partly due to the fact that they do not belong to the natural exponential family. We study the estimation, testing, and variable selection for these models in a unifying framework. The regression models are compared on both synthetic and real RNA-seq data. PMID:28348500

Regression Models For Multivariate Count Data.

PubMed

Zhang, Yiwen; Zhou, Hua; Zhou, Jin; Sun, Wei

2017-01-01

Data with multivariate count responses frequently occur in modern applications. The commonly used multinomial-logit model is limiting due to its restrictive mean-variance structure. For instance, analyzing count data from the recent RNA-seq technology by the multinomial-logit model leads to serious errors in hypothesis testing. The ubiquity of over-dispersion and complicated correlation structures among multivariate counts calls for more flexible regression models. In this article, we study some generalized linear models that incorporate various correlation structures among the counts. Current literature lacks a treatment of these models, partly due to the fact that they do not belong to the natural exponential family. We study the estimation, testing, and variable selection for these models in a unifying framework. The regression models are compared on both synthetic and real RNA-seq data.

Fire spread in chaparral – a comparison of laboratory data and model predictions in burning live fuels

Treesearch

David R. Weise; Eunmo Koo; Xiangyang Zhou; Shankar Mahalingam; Frédéric Morandini; Jacques-Henri Balbi

2016-01-01

Fire behaviour data from 240 laboratory fires in high-density live chaparral fuel beds were compared with model predictions. Logistic regression was used to develop a model to predict fire spread success in the fuel beds and linear regression was used to predict rate of spread. Predictions from the Rothermel equation and three proposed changes as well as two physically...

London Measure of Unplanned Pregnancy: guidance for its use as an outcome measure

PubMed Central

Hall, Jennifer A; Barrett, Geraldine; Copas, Andrew; Stephenson, Judith

2017-01-01

Background The London Measure of Unplanned Pregnancy (LMUP) is a psychometrically validated measure of the degree of intention of a current or recent pregnancy. The LMUP is increasingly being used worldwide, and can be used to evaluate family planning or preconception care programs. However, beyond recommending the use of the full LMUP scale, there is no published guidance on how to use the LMUP as an outcome measure. Ordinal logistic regression has been recommended informally, but studies published to date have all used binary logistic regression and dichotomized the scale at different cut points. There is thus a need for evidence-based guidance to provide a standardized methodology for multivariate analysis and to enable comparison of results. This paper makes recommendations for the regression method for analysis of the LMUP as an outcome measure. Materials and methods Data collected from 4,244 pregnant women in Malawi were used to compare five regression methods: linear, logistic with two cut points, and ordinal logistic with either the full or grouped LMUP score. The recommendations were then tested on the original UK LMUP data. Results There were small but no important differences in the findings across the regression models. Logistic regression resulted in the largest loss of information, and assumptions were violated for the linear and ordinal logistic regression. Consequently, robust standard errors were used for linear regression and a partial proportional odds ordinal logistic regression model attempted. The latter could only be fitted for grouped LMUP score. Conclusion We recommend the linear regression model with robust standard errors to make full use of the LMUP score when analyzed as an outcome measure. Ordinal logistic regression could be considered, but a partial proportional odds model with grouped LMUP score may be required. Logistic regression is the least-favored option, due to the loss of information. For logistic regression, the cut point for un/planned pregnancy should be between nine and ten. These recommendations will standardize the analysis of LMUP data and enhance comparability of results across studies. PMID:28435343

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.

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.

A Comparison between Multiple Regression Models and CUN-BAE Equation to Predict Body Fat in Adults

PubMed Central

Fuster-Parra, Pilar; Bennasar-Veny, Miquel; Tauler, Pedro; Yañez, Aina; López-González, Angel A.; Aguiló, Antoni

2015-01-01

Background Because the accurate measure of body fat (BF) is difficult, several prediction equations have been proposed. The aim of this study was to compare different multiple regression models to predict BF, including the recently reported CUN-BAE equation. Methods Multi regression models using body mass index (BMI) and body adiposity index (BAI) as predictors of BF will be compared. These models will be also compared with the CUN-BAE equation. For all the analysis a sample including all the participants and another one including only the overweight and obese subjects will be considered. The BF reference measure was made using Bioelectrical Impedance Analysis. Results The simplest models including only BMI or BAI as independent variables showed that BAI is a better predictor of BF. However, adding the variable sex to both models made BMI a better predictor than the BAI. For both the whole group of participants and the group of overweight and obese participants, using simple models (BMI, age and sex as variables) allowed obtaining similar correlations with BF as when the more complex CUN-BAE was used (ρ = 0:87 vs. ρ = 0:86 for the whole sample and ρ = 0:88 vs. ρ = 0:89 for overweight and obese subjects, being the second value the one for CUN-BAE). Conclusions There are simpler models than CUN-BAE equation that fits BF as well as CUN-BAE does. Therefore, it could be considered that CUN-BAE overfits. Using a simple linear regression model, the BAI, as the only variable, predicts BF better than BMI. However, when the sex variable is introduced, BMI becomes the indicator of choice to predict BF. PMID:25821960

A comparison between multiple regression models and CUN-BAE equation to predict body fat in adults.

PubMed

Fuster-Parra, Pilar; Bennasar-Veny, Miquel; Tauler, Pedro; Yañez, Aina; López-González, Angel A; Aguiló, Antoni

2015-01-01

Because the accurate measure of body fat (BF) is difficult, several prediction equations have been proposed. The aim of this study was to compare different multiple regression models to predict BF, including the recently reported CUN-BAE equation. Multi regression models using body mass index (BMI) and body adiposity index (BAI) as predictors of BF will be compared. These models will be also compared with the CUN-BAE equation. For all the analysis a sample including all the participants and another one including only the overweight and obese subjects will be considered. The BF reference measure was made using Bioelectrical Impedance Analysis. The simplest models including only BMI or BAI as independent variables showed that BAI is a better predictor of BF. However, adding the variable sex to both models made BMI a better predictor than the BAI. For both the whole group of participants and the group of overweight and obese participants, using simple models (BMI, age and sex as variables) allowed obtaining similar correlations with BF as when the more complex CUN-BAE was used (ρ = 0:87 vs. ρ = 0:86 for the whole sample and ρ = 0:88 vs. ρ = 0:89 for overweight and obese subjects, being the second value the one for CUN-BAE). There are simpler models than CUN-BAE equation that fits BF as well as CUN-BAE does. Therefore, it could be considered that CUN-BAE overfits. Using a simple linear regression model, the BAI, as the only variable, predicts BF better than BMI. However, when the sex variable is introduced, BMI becomes the indicator of choice to predict BF.

Guidelines and Procedures for Computing Time-Series Suspended-Sediment Concentrations and Loads from In-Stream Turbidity-Sensor and Streamflow Data

USGS Publications Warehouse

Rasmussen, Patrick P.; Gray, John R.; Glysson, G. Douglas; Ziegler, Andrew C.

2009-01-01

In-stream continuous turbidity and streamflow data, calibrated with measured suspended-sediment concentration data, can be used to compute a time series of suspended-sediment concentration and load at a stream site. Development of a simple linear (ordinary least squares) regression model for computing suspended-sediment concentrations from instantaneous turbidity data is the first step in the computation process. If the model standard percentage error (MSPE) of the simple linear regression model meets a minimum criterion, this model should be used to compute a time series of suspended-sediment concentrations. Otherwise, a multiple linear regression model using paired instantaneous turbidity and streamflow data is developed and compared to the simple regression model. If the inclusion of the streamflow variable proves to be statistically significant and the uncertainty associated with the multiple regression model results in an improvement over that for the simple linear model, the turbidity-streamflow multiple linear regression model should be used to compute a suspended-sediment concentration time series. The computed concentration time series is subsequently used with its paired streamflow time series to compute suspended-sediment loads by standard U.S. Geological Survey techniques. Once an acceptable regression model is developed, it can be used to compute suspended-sediment concentration beyond the period of record used in model development with proper ongoing collection and analysis of calibration samples. Regression models to compute suspended-sediment concentrations are generally site specific and should never be considered static, but they represent a set period in a continually dynamic system in which additional data will help verify any change in sediment load, type, and source.

Poisson regression models outperform the geometrical model in estimating the peak-to-trough ratio of seasonal variation: a simulation study.

PubMed

Christensen, A L; Lundbye-Christensen, S; Dethlefsen, C

2011-12-01

Several statistical methods of assessing seasonal variation are available. Brookhart and Rothman [3] proposed a second-order moment-based estimator based on the geometrical model derived by Edwards [1], and reported that this estimator is superior in estimating the peak-to-trough ratio of seasonal variation compared with Edwards' estimator with respect to bias and mean squared error. Alternatively, seasonal variation may be modelled using a Poisson regression model, which provides flexibility in modelling the pattern of seasonal variation and adjustments for covariates. Based on a Monte Carlo simulation study three estimators, one based on the geometrical model, and two based on log-linear Poisson regression models, were evaluated in regards to bias and standard deviation (SD). We evaluated the estimators on data simulated according to schemes varying in seasonal variation and presence of a secular trend. All methods and analyses in this paper are available in the R package Peak2Trough[13]. Applying a Poisson regression model resulted in lower absolute bias and SD for data simulated according to the corresponding model assumptions. Poisson regression models had lower bias and SD for data simulated to deviate from the corresponding model assumptions than the geometrical model. This simulation study encourages the use of Poisson regression models in estimating the peak-to-trough ratio of seasonal variation as opposed to the geometrical model. Copyright © 2011 Elsevier Ireland Ltd. All rights reserved.

Grassland and cropland net ecosystem production of the U.S. Great Plains: Regression tree model development and comparative analysis

USGS Publications Warehouse

Wylie, Bruce K.; Howard, Daniel; Dahal, Devendra; Gilmanov, Tagir; Ji, Lei; Zhang, Li; Smith, Kelcy

2016-01-01

This paper presents the methodology and results of two ecological-based net ecosystem production (NEP) regression tree models capable of up scaling measurements made at various flux tower sites throughout the U.S. Great Plains. Separate grassland and cropland NEP regression tree models were trained using various remote sensing data and other biogeophysical data, along with 15 flux towers contributing to the grassland model and 15 flux towers for the cropland model. The models yielded weekly mean daily grassland and cropland NEP maps of the U.S. Great Plains at 250 m resolution for 2000–2008. The grassland and cropland NEP maps were spatially summarized and statistically compared. The results of this study indicate that grassland and cropland ecosystems generally performed as weak net carbon (C) sinks, absorbing more C from the atmosphere than they released from 2000 to 2008. Grasslands demonstrated higher carbon sink potential (139 g C·m−2·year−1) than non-irrigated croplands. A closer look into the weekly time series reveals the C fluctuation through time and space for each land cover type.

Predicting surface fuel models and fuel metrics using lidar and CIR imagery in a dense mixed conifer forest

Treesearch

Marek K. Jakubowksi; Qinghua Guo; Brandon Collins; Scott Stephens; Maggi Kelly

2013-01-01

We compared the ability of several classification and regression algorithms to predict forest stand structure metrics and standard surface fuel models. Our study area spans a dense, topographically complex Sierra Nevada mixed-conifer forest. We used clustering, regression trees, and support vector machine algorithms to analyze high density (average 9 pulses/m

Estimation of Logistic Regression Models in Small Samples. A Simulation Study Using a Weakly Informative Default Prior Distribution

ERIC Educational Resources Information Center

Gordovil-Merino, Amalia; Guardia-Olmos, Joan; Pero-Cebollero, Maribel

2012-01-01

In this paper, we used simulations to compare the performance of classical and Bayesian estimations in logistic regression models using small samples. In the performed simulations, conditions were varied, including the type of relationship between independent and dependent variable values (i.e., unrelated and related values), the type of variable…

Comparison of Cox's Regression Model and Parametric Models in Evaluating the Prognostic Factors for Survival after Liver Transplantation in Shiraz during 2000-2012.

PubMed

Adelian, R; Jamali, J; Zare, N; Ayatollahi, S M T; Pooladfar, G R; Roustaei, N

2015-01-01

Identification of the prognostic factors for survival in patients with liver transplantation is challengeable. Various methods of survival analysis have provided different, sometimes contradictory, results from the same data. To compare Cox's regression model with parametric models for determining the independent factors for predicting adults' and pediatrics' survival after liver transplantation. This study was conducted on 183 pediatric patients and 346 adults underwent liver transplantation in Namazi Hospital, Shiraz, southern Iran. The study population included all patients undergoing liver transplantation from 2000 to 2012. The prognostic factors sex, age, Child class, initial diagnosis of the liver disease, PELD/MELD score, and pre-operative laboratory markers were selected for survival analysis. Among 529 patients, 346 (64.5%) were adult and 183 (34.6%) were pediatric cases. Overall, the lognormal distribution was the best-fitting model for adult and pediatric patients. Age in adults (HR=1.16, p<0.05) and weight (HR=2.68, p<0.01) and Child class B (HR=2.12, p<0.05) in pediatric patients were the most important factors for prediction of survival after liver transplantation. Adult patients younger than the mean age and pediatric patients weighing above the mean and Child class A (compared to those with classes B or C) had better survival. Parametric regression model is a good alternative for the Cox's regression model.

Spatio-temporal water quality mapping from satellite images using geographically and temporally weighted regression

NASA Astrophysics Data System (ADS)

Chu, Hone-Jay; Kong, Shish-Jeng; Chang, Chih-Hua

2018-03-01

The turbidity (TB) of a water body varies with time and space. Water quality is traditionally estimated via linear regression based on satellite images. However, estimating and mapping water quality require a spatio-temporal nonstationary model, while TB mapping necessitates the use of geographically and temporally weighted regression (GTWR) and geographically weighted regression (GWR) models, both of which are more precise than linear regression. Given the temporal nonstationary models for mapping water quality, GTWR offers the best option for estimating regional water quality. Compared with GWR, GTWR provides highly reliable information for water quality mapping, boasts a relatively high goodness of fit, improves the explanation of variance from 44% to 87%, and shows a sufficient space-time explanatory power. The seasonal patterns of TB and the main spatial patterns of TB variability can be identified using the estimated TB maps from GTWR and by conducting an empirical orthogonal function (EOF) analysis.

Cox proportional hazards model of myopic regression for laser in situ keratomileusis flap creation with a femtosecond laser and with a mechanical microkeratome.

PubMed

Lin, Meng-Yin; Chang, David C K; Hsu, Wen-Ming; Wang, I-Jong

2012-06-01

To compare predictive factors for postoperative myopic regression between laser in situ keratomileusis (LASIK) with a femtosecond laser and LASIK with a mechanical microkeratome. Nobel Eye Clinic, Taipei, Taiwan. Retrospective comparative study. Refractive outcomes were recorded 1 day, 1 week, and 1, 3, 6, 9, and 12 months after LASIK. A Cox proportional hazards model was used to evaluate the impact of the 2 flap-creating methods and other covariates on postoperative myopic regression. The femtosecond group comprised 409 eyes and the mechanical microkeratome group, 377 eyes. For both methods, significant predictors for myopic regression after LASIK included preoperative manifest spherical equivalent (P=.0001) and central corneal thickness (P=.027). Laser in situ keratomileusis with a mechanical microkeratome had a higher probability of postoperative myopic regression than LASIK with a femtosecond laser (P=.0002). After adjusting for other covariates in the Cox proportional hazards model, the cumulative risk for myopic regression with a mechanical microkeratome was higher than with a femtosecond laser 12 months postoperatively (P=.0002). With the definition of myopic regression as a myopic shift of 0.50 diopter (D) or more and residual myopia of -0.50 D or less, the risk estimate based on the mean covariates in all eyes in the femtosecond group and mechanical microkeratome group at 12 months was 43.6% and 66.9%, respectively. Laser in situ keratomileusis with a mechanical microkeratome had a higher risk for myopic regression than LASIK with a femtosecond laser through 12 months postoperatively. Copyright © 2012. Published by Elsevier Inc.

Comparing lagged linear correlation, lagged regression, Granger causality, and vector autoregression for uncovering associations in EHR data.

PubMed

Levine, Matthew E; Albers, David J; Hripcsak, George

2016-01-01

Time series analysis methods have been shown to reveal clinical and biological associations in data collected in the electronic health record. We wish to develop reliable high-throughput methods for identifying adverse drug effects that are easy to implement and produce readily interpretable results. To move toward this goal, we used univariate and multivariate lagged regression models to investigate associations between twenty pairs of drug orders and laboratory measurements. Multivariate lagged regression models exhibited higher sensitivity and specificity than univariate lagged regression in the 20 examples, and incorporating autoregressive terms for labs and drugs produced more robust signals in cases of known associations among the 20 example pairings. Moreover, including inpatient admission terms in the model attenuated the signals for some cases of unlikely associations, demonstrating how multivariate lagged regression models' explicit handling of context-based variables can provide a simple way to probe for health-care processes that confound analyses of EHR data.

Fuzzy regression modeling for tool performance prediction and degradation detection.

PubMed

Li, X; Er, M J; Lim, B S; Zhou, J H; Gan, O P; Rutkowski, L

2010-10-01

In this paper, the viability of using Fuzzy-Rule-Based Regression Modeling (FRM) algorithm for tool performance and degradation detection is investigated. The FRM is developed based on a multi-layered fuzzy-rule-based hybrid system with Multiple Regression Models (MRM) embedded into a fuzzy logic inference engine that employs Self Organizing Maps (SOM) for clustering. The FRM converts a complex nonlinear problem to a simplified linear format in order to further increase the accuracy in prediction and rate of convergence. The efficacy of the proposed FRM is tested through a case study - namely to predict the remaining useful life of a ball nose milling cutter during a dry machining process of hardened tool steel with a hardness of 52-54 HRc. A comparative study is further made between four predictive models using the same set of experimental data. It is shown that the FRM is superior as compared with conventional MRM, Back Propagation Neural Networks (BPNN) and Radial Basis Function Networks (RBFN) in terms of prediction accuracy and learning speed.

Predicting ecological flow regime at ungaged sites: A comparison of methods

USGS Publications Warehouse

Murphy, Jennifer C.; Knight, Rodney R.; Wolfe, William J.; Gain, W. Scott

2012-01-01

Nineteen ecologically relevant streamflow characteristics were estimated using published rainfall–runoff and regional regression models for six sites with observed daily streamflow records in Kentucky. The regional regression model produced median estimates closer to the observed median for all but two characteristics. The variability of predictions from both models was generally less than the observed variability. The variability of the predictions from the rainfall–runoff model was greater than that from the regional regression model for all but three characteristics. Eight characteristics predicted by the rainfall–runoff model display positive or negative bias across all six sites; biases are not as pronounced for the regional regression model. Results suggest that a rainfall–runoff model calibrated on a single characteristic is less likely to perform well as a predictor of a range of other characteristics (flow regime) when compared with a regional regression model calibrated individually on multiple characteristics used to represent the flow regime. Poor model performance may misrepresent hydrologic conditions, potentially distorting the perceived risk of ecological degradation. Without prior selection of streamflow characteristics, targeted calibration, and error quantification, the widespread application of general hydrologic models to ecological flow studies is problematic. Published 2012. This article is a U.S. Government work and is in the public domain in the USA.

Prediction model for the return to work of workers with injuries in Hong Kong.

PubMed

Xu, Yanwen; Chan, Chetwyn C H; Lo, Karen Hui Yu-Ling; Tang, Dan

2008-01-01

This study attempts to formulate a prediction model of return to work for a group of workers who have been suffering from chronic pain and physical injury while also being out of work in Hong Kong. The study used Case-based Reasoning (CBR) method, and compared the result with the statistical method of logistic regression model. The database of the algorithm of CBR was composed of 67 cases who were also used in the logistic regression model. The testing cases were 32 participants who had a similar background and characteristics to those in the database. The methods of setting constraints and Euclidean distance metric were used in CBR to search the closest cases to the trial case based on the matrix. The usefulness of the algorithm was tested on 32 new participants, and the accuracy of predicting return to work outcomes was 62.5%, which was no better than the 71.2% accuracy derived from the logistic regression model. The results of the study would enable us to have a better understanding of the CBR applied in the field of occupational rehabilitation by comparing with the conventional regression analysis. The findings would also shed light on the development of relevant interventions for the return-to-work process of these workers.

Evaluation of three statistical prediction models for forensic age prediction based on DNA methylation.

PubMed

Smeers, Inge; Decorte, Ronny; Van de Voorde, Wim; Bekaert, Bram

2018-05-01

DNA methylation is a promising biomarker for forensic age prediction. A challenge that has emerged in recent studies is the fact that prediction errors become larger with increasing age due to interindividual differences in epigenetic ageing rates. This phenomenon of non-constant variance or heteroscedasticity violates an assumption of the often used method of ordinary least squares (OLS) regression. The aim of this study was to evaluate alternative statistical methods that do take heteroscedasticity into account in order to provide more accurate, age-dependent prediction intervals. A weighted least squares (WLS) regression is proposed as well as a quantile regression model. Their performances were compared against an OLS regression model based on the same dataset. Both models provided age-dependent prediction intervals which account for the increasing variance with age, but WLS regression performed better in terms of success rate in the current dataset. However, quantile regression might be a preferred method when dealing with a variance that is not only non-constant, but also not normally distributed. Ultimately the choice of which model to use should depend on the observed characteristics of the data. Copyright © 2018 Elsevier B.V. All rights reserved.

Comparison of Logistic Regression and Artificial Neural Network in Low Back Pain Prediction: Second National Health Survey

PubMed Central

Parsaeian, M; Mohammad, K; Mahmoudi, M; Zeraati, H

2012-01-01

Background: The purpose of this investigation was to compare empirically predictive ability of an artificial neural network with a logistic regression in prediction of low back pain. Methods: Data from the second national health survey were considered in this investigation. This data includes the information of low back pain and its associated risk factors among Iranian people aged 15 years and older. Artificial neural network and logistic regression models were developed using a set of 17294 data and they were validated in a test set of 17295 data. Hosmer and Lemeshow recommendation for model selection was used in fitting the logistic regression. A three-layer perceptron with 9 inputs, 3 hidden and 1 output neurons was employed. The efficiency of two models was compared by receiver operating characteristic analysis, root mean square and -2 Loglikelihood criteria. Results: The area under the ROC curve (SE), root mean square and -2Loglikelihood of the logistic regression was 0.752 (0.004), 0.3832 and 14769.2, respectively. The area under the ROC curve (SE), root mean square and -2Loglikelihood of the artificial neural network was 0.754 (0.004), 0.3770 and 14757.6, respectively. Conclusions: Based on these three criteria, artificial neural network would give better performance than logistic regression. Although, the difference is statistically significant, it does not seem to be clinically significant. PMID:23113198

Comparison of logistic regression and artificial neural network in low back pain prediction: second national health survey.

PubMed

Parsaeian, M; Mohammad, K; Mahmoudi, M; Zeraati, H

2012-01-01

The purpose of this investigation was to compare empirically predictive ability of an artificial neural network with a logistic regression in prediction of low back pain. Data from the second national health survey were considered in this investigation. This data includes the information of low back pain and its associated risk factors among Iranian people aged 15 years and older. Artificial neural network and logistic regression models were developed using a set of 17294 data and they were validated in a test set of 17295 data. Hosmer and Lemeshow recommendation for model selection was used in fitting the logistic regression. A three-layer perceptron with 9 inputs, 3 hidden and 1 output neurons was employed. The efficiency of two models was compared by receiver operating characteristic analysis, root mean square and -2 Loglikelihood criteria. The area under the ROC curve (SE), root mean square and -2Loglikelihood of the logistic regression was 0.752 (0.004), 0.3832 and 14769.2, respectively. The area under the ROC curve (SE), root mean square and -2Loglikelihood of the artificial neural network was 0.754 (0.004), 0.3770 and 14757.6, respectively. Based on these three criteria, artificial neural network would give better performance than logistic regression. Although, the difference is statistically significant, it does not seem to be clinically significant.

Estimation of Covariance Matrix on Bi-Response Longitudinal Data Analysis with Penalized Spline Regression

NASA Astrophysics Data System (ADS)

Islamiyati, A.; Fatmawati; Chamidah, N.

2018-03-01

The correlation assumption of the longitudinal data with bi-response occurs on the measurement between the subjects of observation and the response. It causes the auto-correlation of error, and this can be overcome by using a covariance matrix. In this article, we estimate the covariance matrix based on the penalized spline regression model. Penalized spline involves knot points and smoothing parameters simultaneously in controlling the smoothness of the curve. Based on our simulation study, the estimated regression model of the weighted penalized spline with covariance matrix gives a smaller error value compared to the error of the model without covariance matrix.

Quantile regression in the presence of monotone missingness with sensitivity analysis

PubMed Central

Liu, Minzhao; Daniels, Michael J.; Perri, Michael G.

2016-01-01

In this paper, we develop methods for longitudinal quantile regression when there is monotone missingness. In particular, we propose pattern mixture models with a constraint that provides a straightforward interpretation of the marginal quantile regression parameters. Our approach allows sensitivity analysis which is an essential component in inference for incomplete data. To facilitate computation of the likelihood, we propose a novel way to obtain analytic forms for the required integrals. We conduct simulations to examine the robustness of our approach to modeling assumptions and compare its performance to competing approaches. The model is applied to data from a recent clinical trial on weight management. PMID:26041008

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.

Improving Lidar-based Aboveground Biomass Estimation with Site Productivity for Central Hardwood Forests, USA

NASA Astrophysics Data System (ADS)

Shao, G.; Gallion, J.; Fei, S.

2016-12-01

Sound forest aboveground biomass estimation is required to monitor diverse forest ecosystems and their impacts on the changing climate. Lidar-based regression models provided promised biomass estimations in most forest ecosystems. However, considerable uncertainties of biomass estimations have been reported in the temperate hardwood and hardwood-dominated mixed forests. Varied site productivities in temperate hardwood forests largely diversified height and diameter growth rates, which significantly reduced the correlation between tree height and diameter at breast height (DBH) in mature and complex forests. It is, therefore, difficult to utilize height-based lidar metrics to predict DBH-based field-measured biomass through a simple regression model regardless the variation of site productivity. In this study, we established a multi-dimension nonlinear regression model incorporating lidar metrics and site productivity classes derived from soil features. In the regression model, lidar metrics provided horizontal and vertical structural information and productivity classes differentiated good and poor forest sites. The selection and combination of lidar metrics were discussed. Multiple regression models were employed and compared. Uncertainty analysis was applied to the best fit model. The effects of site productivity on the lidar-based biomass model were addressed.

Electricity Consumption in the Industrial Sector of Jordan: Application of Multivariate Linear Regression and Adaptive Neuro-Fuzzy Techniques

NASA Astrophysics Data System (ADS)

Samhouri, M.; Al-Ghandoor, A.; Fouad, R. H.

2009-08-01

In this study two techniques, for modeling electricity consumption of the Jordanian industrial sector, are presented: (i) multivariate linear regression and (ii) neuro-fuzzy models. Electricity consumption is modeled as function of different variables such as number of establishments, number of employees, electricity tariff, prevailing fuel prices, production outputs, capacity utilizations, and structural effects. It was found that industrial production and capacity utilization are the most important variables that have significant effect on future electrical power demand. The results showed that both the multivariate linear regression and neuro-fuzzy models are generally comparable and can be used adequately to simulate industrial electricity consumption. However, comparison that is based on the square root average squared error of data suggests that the neuro-fuzzy model performs slightly better for future prediction of electricity consumption than the multivariate linear regression model. Such results are in full agreement with similar work, using different methods, for other countries.

On the use of regression analysis for the estimation of human biological age.

PubMed

Krøll, J; Saxtrup, O

2000-01-01

The present investigation compares three linear regression procedures for the definition of human biological age (bioage). As a model system for bioage definition is used the variations with age of blood hemoglobin (B-hemoglobin) in males in the age range 50-95 years. The bioage measures compared are: 1: P-bioage; defined from regression of chronological age on B-hemoglobin results. 2: AC-bioage; obtained by indirect regression, using in reverse the equation describing the regression of B-hemoglobin on age in a reference population. 3: BC-bioage; defined by orthogonal regression on the reference regression line of B-hemoglobin on age. It is demonstrated that the P-bioage measure gives an overestimation of the bioage in the younger and an underestimation in the older individuals. This 'regression to the mean' is avoided using the indirect regression procedures. Here the relatively low SD of the BC-bioage measure results from the inclusion of individual chronological age in the orthogonal regression procedure. Observations on male blood donors illustrates the variation of the AC- and BC-bioage measures in the individual.

Detection of epistatic effects with logic regression and a classical linear regression model.

PubMed

Malina, Magdalena; Ickstadt, Katja; Schwender, Holger; Posch, Martin; Bogdan, Małgorzata

2014-02-01

To locate multiple interacting quantitative trait loci (QTL) influencing a trait of interest within experimental populations, usually methods as the Cockerham's model are applied. Within this framework, interactions are understood as the part of the joined effect of several genes which cannot be explained as the sum of their additive effects. However, if a change in the phenotype (as disease) is caused by Boolean combinations of genotypes of several QTLs, this Cockerham's approach is often not capable to identify them properly. To detect such interactions more efficiently, we propose a logic regression framework. Even though with the logic regression approach a larger number of models has to be considered (requiring more stringent multiple testing correction) the efficient representation of higher order logic interactions in logic regression models leads to a significant increase of power to detect such interactions as compared to a Cockerham's approach. The increase in power is demonstrated analytically for a simple two-way interaction model and illustrated in more complex settings with simulation study and real data analysis.

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.

Bias in logistic regression due to imperfect diagnostic test results and practical correction approaches.

PubMed

Valle, Denis; Lima, Joanna M Tucker; Millar, Justin; Amratia, Punam; Haque, Ubydul

2015-11-04

Logistic regression is a statistical model widely used in cross-sectional and cohort studies to identify and quantify the effects of potential disease risk factors. However, the impact of imperfect tests on adjusted odds ratios (and thus on the identification of risk factors) is under-appreciated. The purpose of this article is to draw attention to the problem associated with modelling imperfect diagnostic tests, and propose simple Bayesian models to adequately address this issue. A systematic literature review was conducted to determine the proportion of malaria studies that appropriately accounted for false-negatives/false-positives in a logistic regression setting. Inference from the standard logistic regression was also compared with that from three proposed Bayesian models using simulations and malaria data from the western Brazilian Amazon. A systematic literature review suggests that malaria epidemiologists are largely unaware of the problem of using logistic regression to model imperfect diagnostic test results. Simulation results reveal that statistical inference can be substantially improved when using the proposed Bayesian models versus the standard logistic regression. Finally, analysis of original malaria data with one of the proposed Bayesian models reveals that microscopy sensitivity is strongly influenced by how long people have lived in the study region, and an important risk factor (i.e., participation in forest extractivism) is identified that would have been missed by standard logistic regression. Given the numerous diagnostic methods employed by malaria researchers and the ubiquitous use of logistic regression to model the results of these diagnostic tests, this paper provides critical guidelines to improve data analysis practice in the presence of misclassification error. Easy-to-use code that can be readily adapted to WinBUGS is provided, enabling straightforward implementation of the proposed Bayesian models.

Comparing the analytical performances of Micro-NIR and FT-NIR spectrometers in the evaluation of acerola fruit quality, using PLS and SVM regression algorithms.

PubMed

Malegori, Cristina; Nascimento Marques, Emanuel José; de Freitas, Sergio Tonetto; Pimentel, Maria Fernanda; Pasquini, Celio; Casiraghi, Ernestina

2017-04-01

The main goal of this study was to investigate the analytical performances of a state-of-the-art device, one of the smallest dispersion NIR spectrometers on the market (MicroNIR 1700), making a critical comparison with a benchtop FT-NIR spectrometer in the evaluation of the prediction accuracy. In particular, the aim of this study was to estimate in a non-destructive manner, titratable acidity and ascorbic acid content in acerola fruit during ripening, in a view of direct applicability in field of this new miniaturised handheld device. Acerola (Malpighia emarginata DC.) is a super-fruit characterised by a considerable amount of ascorbic acid, ranging from 1.0% to 4.5%. However, during ripening, acerola colour changes and the fruit may lose as much as half of its ascorbic acid content. Because the variability of chemical parameters followed a non-strictly linear profile, two different regression algorithms were compared: PLS and SVM. Regression models obtained with Micro-NIR spectra give better results using SVM algorithm, for both ascorbic acid and titratable acidity estimation. FT-NIR data give comparable results using both SVM and PLS algorithms, with lower errors for SVM regression. The prediction ability of the two instruments was statistically compared using the Passing-Bablok regression algorithm; the outcomes are critically discussed together with the regression models, showing the suitability of the portable Micro-NIR for in field monitoring of chemical parameters of interest in acerola fruits. Copyright © 2016 Elsevier B.V. All rights reserved.

Estimating multilevel logistic regression models when the number of clusters is low: a comparison of different statistical software procedures.

PubMed

Austin, Peter C

2010-04-22

Multilevel logistic regression models are increasingly being used to analyze clustered data in medical, public health, epidemiological, and educational research. Procedures for estimating the parameters of such models are available in many statistical software packages. There is currently little evidence on the minimum number of clusters necessary to reliably fit multilevel regression models. We conducted a Monte Carlo study to compare the performance of different statistical software procedures for estimating multilevel logistic regression models when the number of clusters was low. We examined procedures available in BUGS, HLM, R, SAS, and Stata. We found that there were qualitative differences in the performance of different software procedures for estimating multilevel logistic models when the number of clusters was low. Among the likelihood-based procedures, estimation methods based on adaptive Gauss-Hermite approximations to the likelihood (glmer in R and xtlogit in Stata) or adaptive Gaussian quadrature (Proc NLMIXED in SAS) tended to have superior performance for estimating variance components when the number of clusters was small, compared to software procedures based on penalized quasi-likelihood. However, only Bayesian estimation with BUGS allowed for accurate estimation of variance components when there were fewer than 10 clusters. For all statistical software procedures, estimation of variance components tended to be poor when there were only five subjects per cluster, regardless of the number of clusters.

A hybrid PSO-SVM-based method for predicting the friction coefficient between aircraft tire and coating

NASA Astrophysics Data System (ADS)

Zhan, Liwei; Li, Chengwei

2017-02-01

A hybrid PSO-SVM-based model is proposed to predict the friction coefficient between aircraft tire and coating. The presented hybrid model combines a support vector machine (SVM) with particle swarm optimization (PSO) technique. SVM has been adopted to solve regression problems successfully. Its regression accuracy is greatly related to optimizing parameters such as the regularization constant C , the parameter gamma γ corresponding to RBF kernel and the epsilon parameter \\varepsilon in the SVM training procedure. However, the friction coefficient which is predicted based on SVM has yet to be explored between aircraft tire and coating. The experiment reveals that drop height and tire rotational speed are the factors affecting friction coefficient. Bearing in mind, the friction coefficient can been predicted using the hybrid PSO-SVM-based model by the measured friction coefficient between aircraft tire and coating. To compare regression accuracy, a grid search (GS) method and a genetic algorithm (GA) are used to optimize the relevant parameters (C , γ and \\varepsilon ), respectively. The regression accuracy could be reflected by the coefficient of determination ({{R}2} ). The result shows that the hybrid PSO-RBF-SVM-based model has better accuracy compared with the GS-RBF-SVM- and GA-RBF-SVM-based models. The agreement of this model (PSO-RBF-SVM) with experiment data confirms its good performance.

Modeling the frequency of opposing left-turn conflicts at signalized intersections using generalized linear regression models.

PubMed

Zhang, Xin; Liu, Pan; Chen, Yuguang; Bai, Lu; Wang, Wei

2014-01-01

The primary objective of this study was to identify whether the frequency of traffic conflicts at signalized intersections can be modeled. The opposing left-turn conflicts were selected for the development of conflict predictive models. Using data collected at 30 approaches at 20 signalized intersections, the underlying distributions of the conflicts under different traffic conditions were examined. Different conflict-predictive models were developed to relate the frequency of opposing left-turn conflicts to various explanatory variables. The models considered include a linear regression model, a negative binomial model, and separate models developed for four traffic scenarios. The prediction performance of different models was compared. The frequency of traffic conflicts follows a negative binominal distribution. The linear regression model is not appropriate for the conflict frequency data. In addition, drivers behaved differently under different traffic conditions. Accordingly, the effects of conflicting traffic volumes on conflict frequency vary across different traffic conditions. The occurrences of traffic conflicts at signalized intersections can be modeled using generalized linear regression models. The use of conflict predictive models has potential to expand the uses of surrogate safety measures in safety estimation and evaluation.

[Calculating Pearson residual in logistic regressions: a comparison between SPSS and SAS].

PubMed

Xu, Hao; Zhang, Tao; Li, Xiao-song; Liu, Yuan-yuan

2015-01-01

To compare the results of Pearson residual calculations in logistic regression models using SPSS and SAS. We reviewed Pearson residual calculation methods, and used two sets of data to test logistic models constructed by SPSS and STATA. One model contained a small number of covariates compared to the number of observed. The other contained a similar number of covariates as the number of observed. The two software packages produced similar Pearson residual estimates when the models contained a similar number of covariates as the number of observed, but the results differed when the number of observed was much greater than the number of covariates. The two software packages produce different results of Pearson residuals, especially when the models contain a small number of covariates. Further studies are warranted.

A comparison of methods for the analysis of binomial clustered outcomes in behavioral research.

PubMed

Ferrari, Alberto; Comelli, Mario

2016-12-01

In behavioral research, data consisting of a per-subject proportion of "successes" and "failures" over a finite number of trials often arise. This clustered binary data are usually non-normally distributed, which can distort inference if the usual general linear model is applied and sample size is small. A number of more advanced methods is available, but they are often technically challenging and a comparative assessment of their performances in behavioral setups has not been performed. We studied the performances of some methods applicable to the analysis of proportions; namely linear regression, Poisson regression, beta-binomial regression and Generalized Linear Mixed Models (GLMMs). We report on a simulation study evaluating power and Type I error rate of these models in hypothetical scenarios met by behavioral researchers; plus, we describe results from the application of these methods on data from real experiments. Our results show that, while GLMMs are powerful instruments for the analysis of clustered binary outcomes, beta-binomial regression can outperform them in a range of scenarios. Linear regression gave results consistent with the nominal level of significance, but was overall less powerful. Poisson regression, instead, mostly led to anticonservative inference. GLMMs and beta-binomial regression are generally more powerful than linear regression; yet linear regression is robust to model misspecification in some conditions, whereas Poisson regression suffers heavily from violations of the assumptions when used to model proportion data. We conclude providing directions to behavioral scientists dealing with clustered binary data and small sample sizes. Copyright © 2016 Elsevier B.V. All rights reserved.

Building interpretable predictive models for pediatric hospital readmission using Tree-Lasso logistic regression.

PubMed

Jovanovic, Milos; Radovanovic, Sandro; Vukicevic, Milan; Van Poucke, Sven; Delibasic, Boris

2016-09-01

Quantification and early identification of unplanned readmission risk have the potential to improve the quality of care during hospitalization and after discharge. However, high dimensionality, sparsity, and class imbalance of electronic health data and the complexity of risk quantification, challenge the development of accurate predictive models. Predictive models require a certain level of interpretability in order to be applicable in real settings and create actionable insights. This paper aims to develop accurate and interpretable predictive models for readmission in a general pediatric patient population, by integrating a data-driven model (sparse logistic regression) and domain knowledge based on the international classification of diseases 9th-revision clinical modification (ICD-9-CM) hierarchy of diseases. Additionally, we propose a way to quantify the interpretability of a model and inspect the stability of alternative solutions. The analysis was conducted on >66,000 pediatric hospital discharge records from California, State Inpatient Databases, Healthcare Cost and Utilization Project between 2009 and 2011. We incorporated domain knowledge based on the ICD-9-CM hierarchy in a data driven, Tree-Lasso regularized logistic regression model, providing the framework for model interpretation. This approach was compared with traditional Lasso logistic regression resulting in models that are easier to interpret by fewer high-level diagnoses, with comparable prediction accuracy. The results revealed that the use of a Tree-Lasso model was as competitive in terms of accuracy (measured by area under the receiver operating characteristic curve-AUC) as the traditional Lasso logistic regression, but integration with the ICD-9-CM hierarchy of diseases provided more interpretable models in terms of high-level diagnoses. Additionally, interpretations of models are in accordance with existing medical understanding of pediatric readmission. Best performing models have similar performances reaching AUC values 0.783 and 0.779 for traditional Lasso and Tree-Lasso, respectfully. However, information loss of Lasso models is 0.35 bits higher compared to Tree-Lasso model. We propose a method for building predictive models applicable for the detection of readmission risk based on Electronic Health records. Integration of domain knowledge (in the form of ICD-9-CM taxonomy) and a data-driven, sparse predictive algorithm (Tree-Lasso Logistic Regression) resulted in an increase of interpretability of the resulting model. The models are interpreted for the readmission prediction problem in general pediatric population in California, as well as several important subpopulations, and the interpretations of models comply with existing medical understanding of pediatric readmission. Finally, quantitative assessment of the interpretability of the models is given, that is beyond simple counts of selected low-level features. Copyright © 2016 Elsevier B.V. All rights reserved.

Comparative analysis of neural network and regression based condition monitoring approaches for wind turbine fault detection

NASA Astrophysics Data System (ADS)

Schlechtingen, Meik; Ferreira Santos, Ilmar

2011-07-01

This paper presents the research results of a comparison of three different model based approaches for wind turbine fault detection in online SCADA data, by applying developed models to five real measured faults and anomalies. The regression based model as the simplest approach to build a normal behavior model is compared to two artificial neural network based approaches, which are a full signal reconstruction and an autoregressive normal behavior model. Based on a real time series containing two generator bearing damages the capabilities of identifying the incipient fault prior to the actual failure are investigated. The period after the first bearing damage is used to develop the three normal behavior models. The developed or trained models are used to investigate how the second damage manifests in the prediction error. Furthermore the full signal reconstruction and the autoregressive approach are applied to further real time series containing gearbox bearing damages and stator temperature anomalies. The comparison revealed all three models being capable of detecting incipient faults. However, they differ in the effort required for model development and the remaining operational time after first indication of damage. The general nonlinear neural network approaches outperform the regression model. The remaining seasonality in the regression model prediction error makes it difficult to detect abnormality and leads to increased alarm levels and thus a shorter remaining operational period. For the bearing damages and the stator anomalies under investigation the full signal reconstruction neural network gave the best fault visibility and thus led to the highest confidence level.

Third molar development: measurements versus scores as age predictor.

PubMed

Thevissen, P W; Fieuws, S; Willems, G

2011-10-01

Human third molar development is widely used to predict chronological age of sub adult individuals with unknown or doubted age. For these predictions, classically, the radiologically observed third molar growth and maturation is registered using a staging and related scoring technique. Measures of lengths and widths of the developing wisdom tooth and its adjacent second molar can be considered as an alternative registration. The aim of this study was to verify relations between mandibular third molar developmental stages or measurements of mandibular second molar and third molars and age. Age related performance of stages and measurements were compared to assess if measurements added information to age predictions from third molar formation stage. The sample was 340 orthopantomograms (170 females, 170 males) of individuals homogenously distributed in age between 7 and 24 years. Mandibular lower right, third and second molars, were staged following Gleiser and Hunt, length and width measurements were registered, and various ratios of these measurements were calculated. Univariable regression models with age as response and third molar stage, measurements and ratios of second and third molars as predictors, were considered. Multivariable regression models assessed if measurements or ratios added information to age prediction from third molar stage. Coefficients of determination (R(2)) and root mean squared errors (RMSE) obtained from all regression models were compared. The univariable regression model using stages as predictor yielded most accurate age predictions (males: R(2) 0.85, RMSE between 0.85 and 1.22 year; females: R(2) 0.77, RMSE between 1.19 and 2.11 year) compared to all models including measurements and ratios. The multivariable regression models indicated that measurements and ratios added no clinical relevant information to the age prediction from third molar stage. Ratios and measurements of second and third molars are less accurate age predictors than stages of developing third molars. Copyright © 2011 Elsevier Ltd. All rights reserved.

The alarming problems of confounding equivalence using logistic regression models in the perspective of causal diagrams.

PubMed

Yu, Yuanyuan; Li, Hongkai; Sun, Xiaoru; Su, Ping; Wang, Tingting; Liu, Yi; Yuan, Zhongshang; Liu, Yanxun; Xue, Fuzhong

2017-12-28

Confounders can produce spurious associations between exposure and outcome in observational studies. For majority of epidemiologists, adjusting for confounders using logistic regression model is their habitual method, though it has some problems in accuracy and precision. It is, therefore, important to highlight the problems of logistic regression and search the alternative method. Four causal diagram models were defined to summarize confounding equivalence. Both theoretical proofs and simulation studies were performed to verify whether conditioning on different confounding equivalence sets had the same bias-reducing potential and then to select the optimum adjusting strategy, in which logistic regression model and inverse probability weighting based marginal structural model (IPW-based-MSM) were compared. The "do-calculus" was used to calculate the true causal effect of exposure on outcome, then the bias and standard error were used to evaluate the performances of different strategies. Adjusting for different sets of confounding equivalence, as judged by identical Markov boundaries, produced different bias-reducing potential in the logistic regression model. For the sets satisfied G-admissibility, adjusting for the set including all the confounders reduced the equivalent bias to the one containing the parent nodes of the outcome, while the bias after adjusting for the parent nodes of exposure was not equivalent to them. In addition, all causal effect estimations through logistic regression were biased, although the estimation after adjusting for the parent nodes of exposure was nearest to the true causal effect. However, conditioning on different confounding equivalence sets had the same bias-reducing potential under IPW-based-MSM. Compared with logistic regression, the IPW-based-MSM could obtain unbiased causal effect estimation when the adjusted confounders satisfied G-admissibility and the optimal strategy was to adjust for the parent nodes of outcome, which obtained the highest precision. All adjustment strategies through logistic regression were biased for causal effect estimation, while IPW-based-MSM could always obtain unbiased estimation when the adjusted set satisfied G-admissibility. Thus, IPW-based-MSM was recommended to adjust for confounders set.

Inferring epidemiological parameters from phylogenies using regression-ABC: A comparative study

PubMed Central

Gascuel, Olivier

2017-01-01

Inferring epidemiological parameters such as the R0 from time-scaled phylogenies is a timely challenge. Most current approaches rely on likelihood functions, which raise specific issues that range from computing these functions to finding their maxima numerically. Here, we present a new regression-based Approximate Bayesian Computation (ABC) approach, which we base on a large variety of summary statistics intended to capture the information contained in the phylogeny and its corresponding lineage-through-time plot. The regression step involves the Least Absolute Shrinkage and Selection Operator (LASSO) method, which is a robust machine learning technique. It allows us to readily deal with the large number of summary statistics, while avoiding resorting to Markov Chain Monte Carlo (MCMC) techniques. To compare our approach to existing ones, we simulated target trees under a variety of epidemiological models and settings, and inferred parameters of interest using the same priors. We found that, for large phylogenies, the accuracy of our regression-ABC is comparable to that of likelihood-based approaches involving birth-death processes implemented in BEAST2. Our approach even outperformed these when inferring the host population size with a Susceptible-Infected-Removed epidemiological model. It also clearly outperformed a recent kernel-ABC approach when assuming a Susceptible-Infected epidemiological model with two host types. Lastly, by re-analyzing data from the early stages of the recent Ebola epidemic in Sierra Leone, we showed that regression-ABC provides more realistic estimates for the duration parameters (latency and infectiousness) than the likelihood-based method. Overall, ABC based on a large variety of summary statistics and a regression method able to perform variable selection and avoid overfitting is a promising approach to analyze large phylogenies. PMID:28263987

Spatial interpolation schemes of daily precipitation for hydrologic modeling

USGS Publications Warehouse

Hwang, Y.; Clark, M.R.; Rajagopalan, B.; Leavesley, G.

2012-01-01

Distributed hydrologic models typically require spatial estimates of precipitation interpolated from sparsely located observational points to the specific grid points. We compare and contrast the performance of regression-based statistical methods for the spatial estimation of precipitation in two hydrologically different basins and confirmed that widely used regression-based estimation schemes fail to describe the realistic spatial variability of daily precipitation field. The methods assessed are: (1) inverse distance weighted average; (2) multiple linear regression (MLR); (3) climatological MLR; and (4) locally weighted polynomial regression (LWP). In order to improve the performance of the interpolations, the authors propose a two-step regression technique for effective daily precipitation estimation. In this simple two-step estimation process, precipitation occurrence is first generated via a logistic regression model before estimate the amount of precipitation separately on wet days. This process generated the precipitation occurrence, amount, and spatial correlation effectively. A distributed hydrologic model (PRMS) was used for the impact analysis in daily time step simulation. Multiple simulations suggested noticeable differences between the input alternatives generated by three different interpolation schemes. Differences are shown in overall simulation error against the observations, degree of explained variability, and seasonal volumes. Simulated streamflows also showed different characteristics in mean, maximum, minimum, and peak flows. Given the same parameter optimization technique, LWP input showed least streamflow error in Alapaha basin and CMLR input showed least error (still very close to LWP) in Animas basin. All of the two-step interpolation inputs resulted in lower streamflow error compared to the directly interpolated inputs. ?? 2011 Springer-Verlag.

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.

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.

Comparison of regression models for estimation of isometric wrist joint torques using surface electromyography

PubMed Central

2011-01-01

Background Several regression models have been proposed for estimation of isometric joint torque using surface electromyography (SEMG) signals. Common issues related to torque estimation models are degradation of model accuracy with passage of time, electrode displacement, and alteration of limb posture. This work compares the performance of the most commonly used regression models under these circumstances, in order to assist researchers with identifying the most appropriate model for a specific biomedical application. Methods Eleven healthy volunteers participated in this study. A custom-built rig, equipped with a torque sensor, was used to measure isometric torque as each volunteer flexed and extended his wrist. SEMG signals from eight forearm muscles, in addition to wrist joint torque data were gathered during the experiment. Additional data were gathered one hour and twenty-four hours following the completion of the first data gathering session, for the purpose of evaluating the effects of passage of time and electrode displacement on accuracy of models. Acquired SEMG signals were filtered, rectified, normalized and then fed to models for training. Results It was shown that mean adjusted coefficient of determination (Ra2) values decrease between 20%-35% for different models after one hour while altering arm posture decreased mean Ra2 values between 64% to 74% for different models. Conclusions Model estimation accuracy drops significantly with passage of time, electrode displacement, and alteration of limb posture. Therefore model retraining is crucial for preserving estimation accuracy. Data resampling can significantly reduce model training time without losing estimation accuracy. Among the models compared, ordinary least squares linear regression model (OLS) was shown to have high isometric torque estimation accuracy combined with very short training times. PMID:21943179

A comparative study on entrepreneurial attitudes modeled with logistic regression and Bayes nets.

PubMed

López Puga, Jorge; García García, Juan

2012-11-01

Entrepreneurship research is receiving increasing attention in our context, as entrepreneurs are key social agents involved in economic development. We compare the success of the dichotomic logistic regression model and the Bayes simple classifier to predict entrepreneurship, after manipulating the percentage of missing data and the level of categorization in predictors. A sample of undergraduate university students (N = 1230) completed five scales (motivation, attitude towards business creation, obstacles, deficiencies, and training needs) and we found that each of them predicted different aspects of the tendency to business creation. Additionally, our results show that the receiver operating characteristic (ROC) curve is affected by the rate of missing data in both techniques, but logistic regression seems to be more vulnerable when faced with missing data, whereas Bayes nets underperform slightly when categorization has been manipulated. Our study sheds light on the potential entrepreneur profile and we propose to use Bayesian networks as an additional alternative to overcome the weaknesses of logistic regression when missing data are present in applied research.

Modelling of binary logistic regression for obesity among secondary students in a rural area of Kedah

NASA Astrophysics Data System (ADS)

Kamaruddin, Ainur Amira; Ali, Zalila; Noor, Norlida Mohd.; Baharum, Adam; Ahmad, Wan Muhamad Amir W.

2014-07-01

Logistic regression analysis examines the influence of various factors on a dichotomous outcome by estimating the probability of the event's occurrence. Logistic regression, also called a logit model, is a statistical procedure used to model dichotomous outcomes. In the logit model the log odds of the dichotomous outcome is modeled as a linear combination of the predictor variables. The log odds ratio in logistic regression provides a description of the probabilistic relationship of the variables and the outcome. In conducting logistic regression, selection procedures are used in selecting important predictor variables, diagnostics are used to check that assumptions are valid which include independence of errors, linearity in the logit for continuous variables, absence of multicollinearity, and lack of strongly influential outliers and a test statistic is calculated to determine the aptness of the model. This study used the binary logistic regression model to investigate overweight and obesity among rural secondary school students on the basis of their demographics profile, medical history, diet and lifestyle. The results indicate that overweight and obesity of students are influenced by obesity in family and the interaction between a student's ethnicity and routine meals intake. The odds of a student being overweight and obese are higher for a student having a family history of obesity and for a non-Malay student who frequently takes routine meals as compared to a Malay student.

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

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…

Multivariate generalized hidden Markov regression models with random covariates: Physical exercise in an elderly population.

PubMed

Punzo, Antonio; Ingrassia, Salvatore; Maruotti, Antonello

2018-04-22

A time-varying latent variable model is proposed to jointly analyze multivariate mixed-support longitudinal data. The proposal can be viewed as an extension of hidden Markov regression models with fixed covariates (HMRMFCs), which is the state of the art for modelling longitudinal data, with a special focus on the underlying clustering structure. HMRMFCs are inadequate for applications in which a clustering structure can be identified in the distribution of the covariates, as the clustering is independent from the covariates distribution. Here, hidden Markov regression models with random covariates are introduced by explicitly specifying state-specific distributions for the covariates, with the aim of improving the recovering of the clusters in the data with respect to a fixed covariates paradigm. The hidden Markov regression models with random covariates class is defined focusing on the exponential family, in a generalized linear model framework. Model identifiability conditions are sketched, an expectation-maximization algorithm is outlined for parameter estimation, and various implementation and operational issues are discussed. Properties of the estimators of the regression coefficients, as well as of the hidden path parameters, are evaluated through simulation experiments and compared with those of HMRMFCs. The method is applied to physical activity data. Copyright © 2018 John Wiley & Sons, Ltd.

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.

Comparison of Cox’s Regression Model and Parametric Models in Evaluating the Prognostic Factors for Survival after Liver Transplantation in Shiraz during 2000–2012

PubMed Central

Adelian, R.; Jamali, J.; Zare, N.; Ayatollahi, S. M. T.; Pooladfar, G. R.; Roustaei, N.

2015-01-01

Background: Identification of the prognostic factors for survival in patients with liver transplantation is challengeable. Various methods of survival analysis have provided different, sometimes contradictory, results from the same data. Objective: To compare Cox’s regression model with parametric models for determining the independent factors for predicting adults’ and pediatrics’ survival after liver transplantation. Method: This study was conducted on 183 pediatric patients and 346 adults underwent liver transplantation in Namazi Hospital, Shiraz, southern Iran. The study population included all patients undergoing liver transplantation from 2000 to 2012. The prognostic factors sex, age, Child class, initial diagnosis of the liver disease, PELD/MELD score, and pre-operative laboratory markers were selected for survival analysis. Result: Among 529 patients, 346 (64.5%) were adult and 183 (34.6%) were pediatric cases. Overall, the lognormal distribution was the best-fitting model for adult and pediatric patients. Age in adults (HR=1.16, p<0.05) and weight (HR=2.68, p<0.01) and Child class B (HR=2.12, p<0.05) in pediatric patients were the most important factors for prediction of survival after liver transplantation. Adult patients younger than the mean age and pediatric patients weighing above the mean and Child class A (compared to those with classes B or C) had better survival. Conclusion: Parametric regression model is a good alternative for the Cox’s regression model. PMID:26306158

A comparative analysis of predictive models of morbidity in intensive care unit after cardiac surgery - part II: an illustrative example.

PubMed

Cevenini, Gabriele; Barbini, Emanuela; Scolletta, Sabino; Biagioli, Bonizella; Giomarelli, Pierpaolo; Barbini, Paolo

2007-11-22

Popular predictive models for estimating morbidity probability after heart surgery are compared critically in a unitary framework. The study is divided into two parts. In the first part modelling techniques and intrinsic strengths and weaknesses of different approaches were discussed from a theoretical point of view. In this second part the performances of the same models are evaluated in an illustrative example. Eight models were developed: Bayes linear and quadratic models, k-nearest neighbour model, logistic regression model, Higgins and direct scoring systems and two feed-forward artificial neural networks with one and two layers. Cardiovascular, respiratory, neurological, renal, infectious and hemorrhagic complications were defined as morbidity. Training and testing sets each of 545 cases were used. The optimal set of predictors was chosen among a collection of 78 preoperative, intraoperative and postoperative variables by a stepwise procedure. Discrimination and calibration were evaluated by the area under the receiver operating characteristic curve and Hosmer-Lemeshow goodness-of-fit test, respectively. Scoring systems and the logistic regression model required the largest set of predictors, while Bayesian and k-nearest neighbour models were much more parsimonious. In testing data, all models showed acceptable discrimination capacities, however the Bayes quadratic model, using only three predictors, provided the best performance. All models showed satisfactory generalization ability: again the Bayes quadratic model exhibited the best generalization, while artificial neural networks and scoring systems gave the worst results. Finally, poor calibration was obtained when using scoring systems, k-nearest neighbour model and artificial neural networks, while Bayes (after recalibration) and logistic regression models gave adequate results. Although all the predictive models showed acceptable discrimination performance in the example considered, the Bayes and logistic regression models seemed better than the others, because they also had good generalization and calibration. The Bayes quadratic model seemed to be a convincing alternative to the much more usual Bayes linear and logistic regression models. It showed its capacity to identify a minimum core of predictors generally recognized as essential to pragmatically evaluate the risk of developing morbidity after heart surgery.

The use of generalized estimating equations in the analysis of motor vehicle crash data.

PubMed

Hutchings, Caroline B; Knight, Stacey; Reading, James C

2003-01-01

The purpose of this study was to determine if it is necessary to use generalized estimating equations (GEEs) in the analysis of seat belt effectiveness in preventing injuries in motor vehicle crashes. The 1992 Utah crash dataset was used, excluding crash participants where seat belt use was not appropriate (n=93,633). The model used in the 1996 Report to Congress [Report to congress on benefits of safety belts and motorcycle helmets, based on data from the Crash Outcome Data Evaluation System (CODES). National Center for Statistics and Analysis, NHTSA, Washington, DC, February 1996] was analyzed for all occupants with logistic regression, one level of nesting (occupants within crashes), and two levels of nesting (occupants within vehicles within crashes) to compare the use of GEEs with logistic regression. When using one level of nesting compared to logistic regression, 13 of 16 variance estimates changed more than 10%, and eight of 16 parameter estimates changed more than 10%. In addition, three of the independent variables changed from significant to insignificant (alpha=0.05). With the use of two levels of nesting, two of 16 variance estimates and three of 16 parameter estimates changed more than 10% from the variance and parameter estimates in one level of nesting. One of the independent variables changed from insignificant to significant (alpha=0.05) in the two levels of nesting model; therefore, only two of the independent variables changed from significant to insignificant when the logistic regression model was compared to the two levels of nesting model. The odds ratio of seat belt effectiveness in preventing injuries was 12% lower when a one-level nested model was used. Based on these results, we stress the need to use a nested model and GEEs when analyzing motor vehicle crash data.

Hierarchical cluster-based partial least squares regression (HC-PLSR) is an efficient tool for metamodelling of nonlinear dynamic models.

PubMed

Tøndel, Kristin; Indahl, Ulf G; Gjuvsland, Arne B; Vik, Jon Olav; Hunter, Peter; Omholt, Stig W; Martens, Harald

2011-06-01

Deterministic dynamic models of complex biological systems contain a large number of parameters and state variables, related through nonlinear differential equations with various types of feedback. A metamodel of such a dynamic model is a statistical approximation model that maps variation in parameters and initial conditions (inputs) to variation in features of the trajectories of the state variables (outputs) throughout the entire biologically relevant input space. A sufficiently accurate mapping can be exploited both instrumentally and epistemically. Multivariate regression methodology is a commonly used approach for emulating dynamic models. However, when the input-output relations are highly nonlinear or non-monotone, a standard linear regression approach is prone to give suboptimal results. We therefore hypothesised that a more accurate mapping can be obtained by locally linear or locally polynomial regression. We present here a new method for local regression modelling, Hierarchical Cluster-based PLS regression (HC-PLSR), where fuzzy C-means clustering is used to separate the data set into parts according to the structure of the response surface. We compare the metamodelling performance of HC-PLSR with polynomial partial least squares regression (PLSR) and ordinary least squares (OLS) regression on various systems: six different gene regulatory network models with various types of feedback, a deterministic mathematical model of the mammalian circadian clock and a model of the mouse ventricular myocyte function. Our results indicate that multivariate regression is well suited for emulating dynamic models in systems biology. The hierarchical approach turned out to be superior to both polynomial PLSR and OLS regression in all three test cases. The advantage, in terms of explained variance and prediction accuracy, was largest in systems with highly nonlinear functional relationships and in systems with positive feedback loops. HC-PLSR is a promising approach for metamodelling in systems biology, especially for highly nonlinear or non-monotone parameter to phenotype maps. The algorithm can be flexibly adjusted to suit the complexity of the dynamic model behaviour, inviting automation in the metamodelling of complex systems.

Hierarchical Cluster-based Partial Least Squares Regression (HC-PLSR) is an efficient tool for metamodelling of nonlinear dynamic models

PubMed Central

2011-01-01

Background Deterministic dynamic models of complex biological systems contain a large number of parameters and state variables, related through nonlinear differential equations with various types of feedback. A metamodel of such a dynamic model is a statistical approximation model that maps variation in parameters and initial conditions (inputs) to variation in features of the trajectories of the state variables (outputs) throughout the entire biologically relevant input space. A sufficiently accurate mapping can be exploited both instrumentally and epistemically. Multivariate regression methodology is a commonly used approach for emulating dynamic models. However, when the input-output relations are highly nonlinear or non-monotone, a standard linear regression approach is prone to give suboptimal results. We therefore hypothesised that a more accurate mapping can be obtained by locally linear or locally polynomial regression. We present here a new method for local regression modelling, Hierarchical Cluster-based PLS regression (HC-PLSR), where fuzzy C-means clustering is used to separate the data set into parts according to the structure of the response surface. We compare the metamodelling performance of HC-PLSR with polynomial partial least squares regression (PLSR) and ordinary least squares (OLS) regression on various systems: six different gene regulatory network models with various types of feedback, a deterministic mathematical model of the mammalian circadian clock and a model of the mouse ventricular myocyte function. Results Our results indicate that multivariate regression is well suited for emulating dynamic models in systems biology. The hierarchical approach turned out to be superior to both polynomial PLSR and OLS regression in all three test cases. The advantage, in terms of explained variance and prediction accuracy, was largest in systems with highly nonlinear functional relationships and in systems with positive feedback loops. Conclusions HC-PLSR is a promising approach for metamodelling in systems biology, especially for highly nonlinear or non-monotone parameter to phenotype maps. The algorithm can be flexibly adjusted to suit the complexity of the dynamic model behaviour, inviting automation in the metamodelling of complex systems. PMID:21627852

Regression Models and Fuzzy Logic Prediction of TBM Penetration Rate

NASA Astrophysics Data System (ADS)

Minh, Vu Trieu; Katushin, Dmitri; Antonov, Maksim; Veinthal, Renno

2017-03-01

This paper presents statistical analyses of rock engineering properties and the measured penetration rate of tunnel boring machine (TBM) based on the data of an actual project. The aim of this study is to analyze the influence of rock engineering properties including uniaxial compressive strength (UCS), Brazilian tensile strength (BTS), rock brittleness index (BI), the distance between planes of weakness (DPW), and the alpha angle (Alpha) between the tunnel axis and the planes of weakness on the TBM rate of penetration (ROP). Four (4) statistical regression models (two linear and two nonlinear) are built to predict the ROP of TBM. Finally a fuzzy logic model is developed as an alternative method and compared to the four statistical regression models. Results show that the fuzzy logic model provides better estimations and can be applied to predict the TBM performance. The R-squared value (R2) of the fuzzy logic model scores the highest value of 0.714 over the second runner-up of 0.667 from the multiple variables nonlinear regression model.

Partial least squares for efficient models of fecal indicator bacteria on Great Lakes beaches

USGS Publications Warehouse

Brooks, Wesley R.; Fienen, Michael N.; Corsi, Steven R.

2013-01-01

At public beaches, it is now common to mitigate the impact of water-borne pathogens by posting a swimmer's advisory when the concentration of fecal indicator bacteria (FIB) exceeds an action threshold. Since culturing the bacteria delays public notification when dangerous conditions exist, regression models are sometimes used to predict the FIB concentration based on readily-available environmental measurements. It is hard to know which environmental parameters are relevant to predicting FIB concentration, and the parameters are usually correlated, which can hurt the predictive power of a regression model. Here the method of partial least squares (PLS) is introduced to automate the regression modeling process. Model selection is reduced to the process of setting a tuning parameter to control the decision threshold that separates predicted exceedances of the standard from predicted non-exceedances. The method is validated by application to four Great Lakes beaches during the summer of 2010. Performance of the PLS models compares favorably to that of the existing state-of-the-art regression models at these four sites.

Estimation Methods for Non-Homogeneous Regression - Minimum CRPS vs Maximum Likelihood

NASA Astrophysics Data System (ADS)

Gebetsberger, Manuel; Messner, Jakob W.; Mayr, Georg J.; Zeileis, Achim

2017-04-01

Non-homogeneous regression models are widely used to statistically post-process numerical weather prediction models. Such regression models correct for errors in mean and variance and are capable to forecast a full probability distribution. In order to estimate the corresponding regression coefficients, CRPS minimization is performed in many meteorological post-processing studies since the last decade. In contrast to maximum likelihood estimation, CRPS minimization is claimed to yield more calibrated forecasts. Theoretically, both scoring rules used as an optimization score should be able to locate a similar and unknown optimum. Discrepancies might result from a wrong distributional assumption of the observed quantity. To address this theoretical concept, this study compares maximum likelihood and minimum CRPS estimation for different distributional assumptions. First, a synthetic case study shows that, for an appropriate distributional assumption, both estimation methods yield to similar regression coefficients. The log-likelihood estimator is slightly more efficient. A real world case study for surface temperature forecasts at different sites in Europe confirms these results but shows that surface temperature does not always follow the classical assumption of a Gaussian distribution. KEYWORDS: ensemble post-processing, maximum likelihood estimation, CRPS minimization, probabilistic temperature forecasting, distributional regression models

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.

A Predictive Model for Microbial Counts on Beaches where Intertidal Sand is the Primary Source

PubMed Central

Feng, Zhixuan; Reniers, Ad; Haus, Brian K.; Solo-Gabriele, Helena M.; Wang, John D.; Fleming, Lora E.

2015-01-01

Human health protection at recreational beaches requires accurate and timely information on microbiological conditions to issue advisories. The objective of this study was to develop a new numerical mass balance model for enterococci levels on nonpoint source beaches. The significant advantage of this model is its easy implementation, and it provides a detailed description of the cross-shore distribution of enterococci that is useful for beach management purposes. The performance of the balance model was evaluated by comparing predicted exceedances of a beach advisory threshold value to field data, and to a traditional regression model. Both the balance model and regression equation predicted approximately 70% the advisories correctly at the knee depth and over 90% at the waist depth. The balance model has the advantage over the regression equation in its ability to simulate spatiotemporal variations of microbial levels, and it is recommended for making more informed management decisions. PMID:25840869

Testing homogeneity in Weibull-regression models.

PubMed

Bolfarine, Heleno; Valença, Dione M

2005-10-01

In survival studies with families or geographical units it may be of interest testing whether such groups are homogeneous for given explanatory variables. In this paper we consider score type tests for group homogeneity based on a mixing model in which the group effect is modelled as a random variable. As opposed to hazard-based frailty models, this model presents survival times that conditioned on the random effect, has an accelerated failure time representation. The test statistics requires only estimation of the conventional regression model without the random effect and does not require specifying the distribution of the random effect. The tests are derived for a Weibull regression model and in the uncensored situation, a closed form is obtained for the test statistic. A simulation study is used for comparing the power of the tests. The proposed tests are applied to real data sets with censored data.

Deep ensemble learning of sparse regression models for brain disease diagnosis.

PubMed

Suk, Heung-Il; Lee, Seong-Whan; Shen, Dinggang

2017-04-01

Recent studies on brain imaging analysis witnessed the core roles of machine learning techniques in computer-assisted intervention for brain disease diagnosis. Of various machine-learning techniques, sparse regression models have proved their effectiveness in handling high-dimensional data but with a small number of training samples, especially in medical problems. In the meantime, deep learning methods have been making great successes by outperforming the state-of-the-art performances in various applications. In this paper, we propose a novel framework that combines the two conceptually different methods of sparse regression and deep learning for Alzheimer's disease/mild cognitive impairment diagnosis and prognosis. Specifically, we first train multiple sparse regression models, each of which is trained with different values of a regularization control parameter. Thus, our multiple sparse regression models potentially select different feature subsets from the original feature set; thereby they have different powers to predict the response values, i.e., clinical label and clinical scores in our work. By regarding the response values from our sparse regression models as target-level representations, we then build a deep convolutional neural network for clinical decision making, which thus we call 'Deep Ensemble Sparse Regression Network.' To our best knowledge, this is the first work that combines sparse regression models with deep neural network. In our experiments with the ADNI cohort, we validated the effectiveness of the proposed method by achieving the highest diagnostic accuracies in three classification tasks. We also rigorously analyzed our results and compared with the previous studies on the ADNI cohort in the literature. Copyright © 2017 Elsevier B.V. All rights reserved.

Deep ensemble learning of sparse regression models for brain disease diagnosis

PubMed Central

Suk, Heung-Il; Lee, Seong-Whan; Shen, Dinggang

2018-01-01

Recent studies on brain imaging analysis witnessed the core roles of machine learning techniques in computer-assisted intervention for brain disease diagnosis. Of various machine-learning techniques, sparse regression models have proved their effectiveness in handling high-dimensional data but with a small number of training samples, especially in medical problems. In the meantime, deep learning methods have been making great successes by outperforming the state-of-the-art performances in various applications. In this paper, we propose a novel framework that combines the two conceptually different methods of sparse regression and deep learning for Alzheimer’s disease/mild cognitive impairment diagnosis and prognosis. Specifically, we first train multiple sparse regression models, each of which is trained with different values of a regularization control parameter. Thus, our multiple sparse regression models potentially select different feature subsets from the original feature set; thereby they have different powers to predict the response values, i.e., clinical label and clinical scores in our work. By regarding the response values from our sparse regression models as target-level representations, we then build a deep convolutional neural network for clinical decision making, which thus we call ‘ Deep Ensemble Sparse Regression Network.’ To our best knowledge, this is the first work that combines sparse regression models with deep neural network. In our experiments with the ADNI cohort, we validated the effectiveness of the proposed method by achieving the highest diagnostic accuracies in three classification tasks. We also rigorously analyzed our results and compared with the previous studies on the ADNI cohort in the literature. PMID:28167394

A primer for biomedical scientists on how to execute model II linear regression analysis.

PubMed

Ludbrook, John

2012-04-01

1. There are two very different ways of executing linear regression analysis. One is Model I, when the x-values are fixed by the experimenter. The other is Model II, in which the x-values are free to vary and are subject to error. 2. I have received numerous complaints from biomedical scientists that they have great difficulty in executing Model II linear regression analysis. This may explain the results of a Google Scholar search, which showed that the authors of articles in journals of physiology, pharmacology and biochemistry rarely use Model II regression analysis. 3. I repeat my previous arguments in favour of using least products linear regression analysis for Model II regressions. I review three methods for executing ordinary least products (OLP) and weighted least products (WLP) regression analysis: (i) scientific calculator and/or computer spreadsheet; (ii) specific purpose computer programs; and (iii) general purpose computer programs. 4. Using a scientific calculator and/or computer spreadsheet, it is easy to obtain correct values for OLP slope and intercept, but the corresponding 95% confidence intervals (CI) are inaccurate. 5. Using specific purpose computer programs, the freeware computer program smatr gives the correct OLP regression coefficients and obtains 95% CI by bootstrapping. In addition, smatr can be used to compare the slopes of OLP lines. 6. When using general purpose computer programs, I recommend the commercial programs systat and Statistica for those who regularly undertake linear regression analysis and I give step-by-step instructions in the Supplementary Information as to how to use loss functions. © 2011 The Author. Clinical and Experimental Pharmacology and Physiology. © 2011 Blackwell Publishing Asia Pty Ltd.

Comparison of Conventional and ANN Models for River Flow Forecasting

NASA Astrophysics Data System (ADS)

Jain, A.; Ganti, R.

2011-12-01

Hydrological models are useful in many water resources applications such as flood control, irrigation and drainage, hydro power generation, water supply, erosion and sediment control, etc. Estimates of runoff are needed in many water resources planning, design development, operation and maintenance activities. River flow is generally estimated using time series or rainfall-runoff models. Recently, soft artificial intelligence tools such as Artificial Neural Networks (ANNs) have become popular for research purposes but have not been extensively adopted in operational hydrological forecasts. There is a strong need to develop ANN models based on real catchment data and compare them with the conventional models. In this paper, a comparative study has been carried out for river flow forecasting using the conventional and ANN models. Among the conventional models, multiple linear, and non linear regression, and time series models of auto regressive (AR) type have been developed. Feed forward neural network model structure trained using the back propagation algorithm, a gradient search method, was adopted. The daily river flow data derived from Godavari Basin @ Polavaram, Andhra Pradesh, India have been employed to develop all the models included here. Two inputs, flows at two past time steps, (Q(t-1) and Q(t-2)) were selected using partial auto correlation analysis for forecasting flow at time t, Q(t). A wide range of error statistics have been used to evaluate the performance of all the models developed in this study. It has been found that the regression and AR models performed comparably, and the ANN model performed the best amongst all the models investigated in this study. It is concluded that ANN model should be adopted in real catchments for hydrological modeling and forecasting.

Modeling energy expenditure in children and adolescents using quantile regression

PubMed Central

Yang, Yunwen; Adolph, Anne L.; Puyau, Maurice R.; Vohra, Firoz A.; Zakeri, Issa F.

2013-01-01

Advanced mathematical models have the potential to capture the complex metabolic and physiological processes that result in energy expenditure (EE). Study objective is to apply quantile regression (QR) to predict EE and determine quantile-dependent variation in covariate effects in nonobese and obese children. First, QR models will be developed to predict minute-by-minute awake EE at different quantile levels based on heart rate (HR) and physical activity (PA) accelerometry counts, and child characteristics of age, sex, weight, and height. Second, the QR models will be used to evaluate the covariate effects of weight, PA, and HR across the conditional EE distribution. QR and ordinary least squares (OLS) regressions are estimated in 109 children, aged 5–18 yr. QR modeling of EE outperformed OLS regression for both nonobese and obese populations. Average prediction errors for QR compared with OLS were not only smaller at the median τ = 0.5 (18.6 vs. 21.4%), but also substantially smaller at the tails of the distribution (10.2 vs. 39.2% at τ = 0.1 and 8.7 vs. 19.8% at τ = 0.9). Covariate effects of weight, PA, and HR on EE for the nonobese and obese children differed across quantiles (P < 0.05). The associations (linear and quadratic) between PA and HR with EE were stronger for the obese than nonobese population (P < 0.05). In conclusion, QR provided more accurate predictions of EE compared with conventional OLS regression, especially at the tails of the distribution, and revealed substantially different covariate effects of weight, PA, and HR on EE in nonobese and obese children. PMID:23640591

Regression mixture models: Does modeling the covariance between independent variables and latent classes improve the results?

PubMed Central

Lamont, Andrea E.; Vermunt, Jeroen K.; Van Horn, M. Lee

2016-01-01

Regression mixture models are increasingly used as an exploratory approach to identify heterogeneity in the effects of a predictor on an outcome. In this simulation study, we test the effects of violating an implicit assumption often made in these models – i.e., independent variables in the model are not directly related to latent classes. Results indicated that the major risk of failing to model the relationship between predictor and latent class was an increase in the probability of selecting additional latent classes and biased class proportions. Additionally, this study tests whether regression mixture models can detect a piecewise relationship between a predictor and outcome. Results suggest that these models are able to detect piecewise relations, but only when the relationship between the latent class and the predictor is included in model estimation. We illustrate the implications of making this assumption through a re-analysis of applied data examining heterogeneity in the effects of family resources on academic achievement. We compare previous results (which assumed no relation between independent variables and latent class) to the model where this assumption is lifted. Implications and analytic suggestions for conducting regression mixture based on these findings are noted. PMID:26881956

Marginalized zero-inflated negative binomial regression with application to dental caries

PubMed Central

Preisser, John S.; Das, Kalyan; Long, D. Leann; Divaris, Kimon

2015-01-01

The zero-inflated negative binomial regression model (ZINB) is often employed in diverse fields such as dentistry, health care utilization, highway safety, and medicine to examine relationships between exposures of interest and overdispersed count outcomes exhibiting many zeros. The regression coefficients of ZINB have latent class interpretations for a susceptible subpopulation at risk for the disease/condition under study with counts generated from a negative binomial distribution and for a non-susceptible subpopulation that provides only zero counts. The ZINB parameters, however, are not well-suited for estimating overall exposure effects, specifically, in quantifying the effect of an explanatory variable in the overall mixture population. In this paper, a marginalized zero-inflated negative binomial regression (MZINB) model for independent responses is proposed to model the population marginal mean count directly, providing straightforward inference for overall exposure effects based on maximum likelihood estimation. Through simulation studies, the finite sample performance of MZINB is compared to marginalized zero-inflated Poisson, Poisson, and negative binomial regression. The MZINB model is applied in the evaluation of a school-based fluoride mouthrinse program on dental caries in 677 children. PMID:26568034

Automated time series forecasting for biosurveillance.

PubMed

Burkom, Howard S; Murphy, Sean Patrick; Shmueli, Galit

2007-09-30

For robust detection performance, traditional control chart monitoring for biosurveillance is based on input data free of trends, day-of-week effects, and other systematic behaviour. Time series forecasting methods may be used to remove this behaviour by subtracting forecasts from observations to form residuals for algorithmic input. We describe three forecast methods and compare their predictive accuracy on each of 16 authentic syndromic data streams. The methods are (1) a non-adaptive regression model using a long historical baseline, (2) an adaptive regression model with a shorter, sliding baseline, and (3) the Holt-Winters method for generalized exponential smoothing. Criteria for comparing the forecasts were the root-mean-square error, the median absolute per cent error (MedAPE), and the median absolute deviation. The median-based criteria showed best overall performance for the Holt-Winters method. The MedAPE measures over the 16 test series averaged 16.5, 11.6, and 9.7 for the non-adaptive regression, adaptive regression, and Holt-Winters methods, respectively. The non-adaptive regression forecasts were degraded by changes in the data behaviour in the fixed baseline period used to compute model coefficients. The mean-based criterion was less conclusive because of the effects of poor forecasts on a small number of calendar holidays. The Holt-Winters method was also most effective at removing serial autocorrelation, with most 1-day-lag autocorrelation coefficients below 0.15. The forecast methods were compared without tuning them to the behaviour of individual series. We achieved improved predictions with such tuning of the Holt-Winters method, but practical use of such improvements for routine surveillance will require reliable data classification methods.

Prescription-drug-related risk in driving: comparing conventional and lasso shrinkage logistic regressions.

PubMed

Avalos, Marta; Adroher, Nuria Duran; Lagarde, Emmanuel; Thiessard, Frantz; Grandvalet, Yves; Contrand, Benjamin; Orriols, Ludivine

2012-09-01

Large data sets with many variables provide particular challenges when constructing analytic models. Lasso-related methods provide a useful tool, although one that remains unfamiliar to most epidemiologists. We illustrate the application of lasso methods in an analysis of the impact of prescribed drugs on the risk of a road traffic crash, using a large French nationwide database (PLoS Med 2010;7:e1000366). In the original case-control study, the authors analyzed each exposure separately. We use the lasso method, which can simultaneously perform estimation and variable selection in a single model. We compare point estimates and confidence intervals using (1) a separate logistic regression model for each drug with a Bonferroni correction and (2) lasso shrinkage logistic regression analysis. Shrinkage regression had little effect on (bias corrected) point estimates, but led to less conservative results, noticeably for drugs with moderate levels of exposure. Carbamates, carboxamide derivative and fatty acid derivative antiepileptics, drugs used in opioid dependence, and mineral supplements of potassium showed stronger associations. Lasso is a relevant method in the analysis of databases with large number of exposures and can be recommended as an alternative to conventional strategies.

Application of near-infrared spectroscopy in the detection of fat-soluble vitamins in premix feed

NASA Astrophysics Data System (ADS)

Jia, Lian Ping; Tian, Shu Li; Zheng, Xue Cong; Jiao, Peng; Jiang, Xun Peng

2018-02-01

Vitamin is the organic compound and necessary for animal physiological maintenance. The rapid determination of the content of different vitamins in premix feed can help to achieve accurate diets and efficient feeding. Compared with high-performance liquid chromatography and other wet chemical methods, near-infrared spectroscopy is a fast, non-destructive, non-polluting method. 168 samples of premix feed were collected and the contents of vitamin A, vitamin E and vitamin D3 were detected by the standard method. The near-infrared spectra of samples ranging from 10 000 to 4 000 cm-1 were obtained. Partial least squares regression (PLSR) and support vector machine regression (SVMR) were used to construct the quantitative model. The results showed that the RMSEP of PLSR model of vitamin A, vitamin E and vitamin D3 were 0.43×107 IU/kg, 0.09×105 IU/kg and 0.17×107 IU/kg, respectively. The RMSEP of SVMR model was 0.45×107 IU/kg, 0.11×105 IU/kg and 0.18×107 IU/kg. Compared with nonlinear regression method (SVMR), linear regression method (PLSR) is more suitable for the quantitative analysis of vitamins in premix feed.

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.

Factor complexity of crash occurrence: An empirical demonstration using boosted regression trees.

PubMed

Chung, Yi-Shih

2013-12-01

Factor complexity is a characteristic of traffic crashes. This paper proposes a novel method, namely boosted regression trees (BRT), to investigate the complex and nonlinear relationships in high-variance traffic crash data. The Taiwanese 2004-2005 single-vehicle motorcycle crash data are used to demonstrate the utility of BRT. Traditional logistic regression and classification and regression tree (CART) models are also used to compare their estimation results and external validities. Both the in-sample cross-validation and out-of-sample validation results show that an increase in tree complexity provides improved, although declining, classification performance, indicating a limited factor complexity of single-vehicle motorcycle crashes. The effects of crucial variables including geographical, time, and sociodemographic factors explain some fatal crashes. Relatively unique fatal crashes are better approximated by interactive terms, especially combinations of behavioral factors. BRT models generally provide improved transferability than conventional logistic regression and CART models. This study also discusses the implications of the results for devising safety policies. Copyright © 2012 Elsevier Ltd. All rights reserved.

Simultaneous Force Regression and Movement Classification of Fingers via Surface EMG within a Unified Bayesian Framework.

PubMed

Baldacchino, Tara; Jacobs, William R; Anderson, Sean R; Worden, Keith; Rowson, Jennifer

2018-01-01

This contribution presents a novel methodology for myolectric-based control using surface electromyographic (sEMG) signals recorded during finger movements. A multivariate Bayesian mixture of experts (MoE) model is introduced which provides a powerful method for modeling force regression at the fingertips, while also performing finger movement classification as a by-product of the modeling algorithm. Bayesian inference of the model allows uncertainties to be naturally incorporated into the model structure. This method is tested using data from the publicly released NinaPro database which consists of sEMG recordings for 6 degree-of-freedom force activations for 40 intact subjects. The results demonstrate that the MoE model achieves similar performance compared to the benchmark set by the authors of NinaPro for finger force regression. Additionally, inherent to the Bayesian framework is the inclusion of uncertainty in the model parameters, naturally providing confidence bounds on the force regression predictions. Furthermore, the integrated clustering step allows a detailed investigation into classification of the finger movements, without incurring any extra computational effort. Subsequently, a systematic approach to assessing the importance of the number of electrodes needed for accurate control is performed via sensitivity analysis techniques. A slight degradation in regression performance is observed for a reduced number of electrodes, while classification performance is unaffected.

Optimization of Regression Models of Experimental Data Using Confirmation Points

NASA Technical Reports Server (NTRS)

Ulbrich, N.

2010-01-01

A new search metric is discussed that may be used to better assess the predictive capability of different math term combinations during the optimization of a regression model of experimental data. The new search metric can be determined for each tested math term combination if the given experimental data set is split into two subsets. The first subset consists of data points that are only used to determine the coefficients of the regression model. The second subset consists of confirmation points that are exclusively used to test the regression model. The new search metric value is assigned after comparing two values that describe the quality of the fit of each subset. The first value is the standard deviation of the PRESS residuals of the data points. The second value is the standard deviation of the response residuals of the confirmation points. The greater of the two values is used as the new search metric value. This choice guarantees that both standard deviations are always less or equal to the value that is used during the optimization. Experimental data from the calibration of a wind tunnel strain-gage balance is used to illustrate the application of the new search metric. The new search metric ultimately generates an optimized regression model that was already tested at regression model independent confirmation points before it is ever used to predict an unknown response from a set of regressors.

A land use regression model for ambient ultrafine particles in Montreal, Canada: A comparison of linear regression and a machine learning approach.

PubMed

Weichenthal, Scott; Ryswyk, Keith Van; Goldstein, Alon; Bagg, Scott; Shekkarizfard, Maryam; Hatzopoulou, Marianne

2016-04-01

Existing evidence suggests that ambient ultrafine particles (UFPs) (<0.1µm) may contribute to acute cardiorespiratory morbidity. However, few studies have examined the long-term health effects of these pollutants owing in part to a need for exposure surfaces that can be applied in large population-based studies. To address this need, we developed a land use regression model for UFPs in Montreal, Canada using mobile monitoring data collected from 414 road segments during the summer and winter months between 2011 and 2012. Two different approaches were examined for model development including standard multivariable linear regression and a machine learning approach (kernel-based regularized least squares (KRLS)) that learns the functional form of covariate impacts on ambient UFP concentrations from the data. The final models included parameters for population density, ambient temperature and wind speed, land use parameters (park space and open space), length of local roads and rail, and estimated annual average NOx emissions from traffic. The final multivariable linear regression model explained 62% of the spatial variation in ambient UFP concentrations whereas the KRLS model explained 79% of the variance. The KRLS model performed slightly better than the linear regression model when evaluated using an external dataset (R(2)=0.58 vs. 0.55) or a cross-validation procedure (R(2)=0.67 vs. 0.60). In general, our findings suggest that the KRLS approach may offer modest improvements in predictive performance compared to standard multivariable linear regression models used to estimate spatial variations in ambient UFPs. However, differences in predictive performance were not statistically significant when evaluated using the cross-validation procedure. Crown Copyright © 2015. Published by Elsevier Inc. All rights reserved.

A test of inflated zeros for Poisson regression models.

PubMed

He, Hua; Zhang, Hui; Ye, Peng; Tang, Wan

2017-01-01

Excessive zeros are common in practice and may cause overdispersion and invalidate inference when fitting Poisson regression models. There is a large body of literature on zero-inflated Poisson models. However, methods for testing whether there are excessive zeros are less well developed. The Vuong test comparing a Poisson and a zero-inflated Poisson model is commonly applied in practice. However, the type I error of the test often deviates seriously from the nominal level, rendering serious doubts on the validity of the test in such applications. In this paper, we develop a new approach for testing inflated zeros under the Poisson model. Unlike the Vuong test for inflated zeros, our method does not require a zero-inflated Poisson model to perform the test. Simulation studies show that when compared with the Vuong test our approach not only better at controlling type I error rate, but also yield more power.

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.

Predicting nitrogen loading with land-cover composition: how can watershed size affect model performance?

PubMed

Zhang, Tao; Yang, Xiaojun

2013-01-01

Watershed-wide land-cover proportions can be used to predict the in-stream non-point source pollutant loadings through regression modeling. However, the model performance can vary greatly across different study sites and among various watersheds. Existing literature has shown that this type of regression modeling tends to perform better for large watersheds than for small ones, and that such a performance variation has been largely linked with different interwatershed landscape heterogeneity levels. The purpose of this study is to further examine the previously mentioned empirical observation based on a set of watersheds in the northern part of Georgia (USA) to explore the underlying causes of the variation in model performance. Through the combined use of the neutral landscape modeling approach and a spatially explicit nutrient loading model, we tested whether the regression model performance variation over the watershed groups ranging in size is due to the different watershed landscape heterogeneity levels. We adopted three neutral landscape modeling criteria that were tied with different similarity levels in watershed landscape properties and used the nutrient loading model to estimate the nitrogen loads for these neutral watersheds. Then we compared the regression model performance for the real and neutral landscape scenarios, respectively. We found that watershed size can affect the regression model performance both directly and indirectly. Along with the indirect effect through interwatershed heterogeneity, watershed size can directly affect the model performance over the watersheds varying in size. We also found that the regression model performance can be more significantly affected by other physiographic properties shaping nitrogen delivery effectiveness than the watershed land-cover heterogeneity. This study contrasts with many existing studies because it goes beyond hypothesis formulation based on empirical observations and into hypothesis testing to explore the fundamental mechanism.

Efficient least angle regression for identification of linear-in-the-parameters models

PubMed Central

Beach, Thomas H.; Rezgui, Yacine

2017-01-01

Least angle regression, as a promising model selection method, differentiates itself from conventional stepwise and stagewise methods, in that it is neither too greedy nor too slow. It is closely related to L1 norm optimization, which has the advantage of low prediction variance through sacrificing part of model bias property in order to enhance model generalization capability. In this paper, we propose an efficient least angle regression algorithm for model selection for a large class of linear-in-the-parameters models with the purpose of accelerating the model selection process. The entire algorithm works completely in a recursive manner, where the correlations between model terms and residuals, the evolving directions and other pertinent variables are derived explicitly and updated successively at every subset selection step. The model coefficients are only computed when the algorithm finishes. The direct involvement of matrix inversions is thereby relieved. A detailed computational complexity analysis indicates that the proposed algorithm possesses significant computational efficiency, compared with the original approach where the well-known efficient Cholesky decomposition is involved in solving least angle regression. Three artificial and real-world examples are employed to demonstrate the effectiveness, efficiency and numerical stability of the proposed algorithm. PMID:28293140

ERT to aid in WSN based early warning system for landslides

NASA Astrophysics Data System (ADS)

T, H.

2017-12-01

Amrita University's landslide monitoring and early warning system using Wireless Sensor Networks (WSN) consists of heterogeneous sensors like rain gauge, moisture sensor, piezometer, geophone, inclinometer, tilt meter etc. The information from the sensors are accurate and limited to that point. In order to monitor a large area, ERT can be used in conjunction with WSN technology. To accomplish the feasibility of ERT in landslide early warning along with WSN technology, we have conducted experiments in Amrita's landslide laboratory setup. The experiment was aimed to simulate landslide, and monitor the changes happening in the soil using moisture sensor and ERT. Simulating moisture values from resistivity measurements to a greater accuracy can help in landslide monitoring for large areas. For accomplishing the same we have adapted two mathematical approaches, 1) Regression analysis between resistivity measurements and actual moisture values from moisture sensor, and 2) Using Waxman Smith model to simulate moisture values from resistivity measurements. The simulated moisture values from Waxman Smith model is compared with the actual moisture values and the Mean Square Error (MSE) is found to be 46.33. Regression curve is drawn for the resistivity vs simulated moisture values from Waxman model, and it is compared with the regression curve of actual model, which is shown in figure-1. From figure-1, it is clear that there the regression curve from actual moisture values and the regression curve from simulated moisture values, follow the similar pattern and there is a small difference between them. Moisture values can be simulated to a greater accuracy using actual regression equation, but the limitation is that, regression curves will differ for different sites and different soils. Regression equation from actual moisture values can be used, if we have conducted experiment in the laboratory for a particular soil sample, otherwise with the knowledge of soil properties, Waxman model can be used to simulate moisture values. The promising results assure that, ERT measurements when used in conjunction with WSN technique, vital paramters triggering landslides like moisture can be simulated for a large area, which will help in providing early warning for large areas.

A model of the human in a cognitive prediction task.

NASA Technical Reports Server (NTRS)

Rouse, W. B.

1973-01-01

The human decision maker's behavior when predicting future states of discrete linear dynamic systems driven by zero-mean Gaussian processes is modeled. The task is on a slow enough time scale that physiological constraints are insignificant compared with cognitive limitations. The model is basically a linear regression system identifier with a limited memory and noisy observations. Experimental data are presented and compared to the model.

Creep analysis of silicone for podiatry applications.

PubMed

Janeiro-Arocas, Julia; Tarrío-Saavedra, Javier; López-Beceiro, Jorge; Naya, Salvador; López-Canosa, Adrián; Heredia-García, Nicolás; Artiaga, Ramón

2016-10-01

This work shows an effective methodology to characterize the creep-recovery behavior of silicones before their application in podiatry. The aim is to characterize, model and compare the creep-recovery properties of different types of silicone used in podiatry orthotics. Creep-recovery phenomena of silicones used in podiatry orthotics is characterized by dynamic mechanical analysis (DMA). Silicones provided by Herbitas are compared by observing their viscoelastic properties by Functional Data Analysis (FDA) and nonlinear regression. The relationship between strain and time is modeled by fixed and mixed effects nonlinear regression to compare easily and intuitively podiatry silicones. Functional ANOVA and Kohlrausch-Willians-Watts (KWW) model with fixed and mixed effects allows us to compare different silicones observing the values of fitting parameters and their physical meaning. The differences between silicones are related to the variations of breadth of creep-recovery time distribution and instantaneous deformation-permanent strain. Nevertheless, the mean creep-relaxation time is the same for all the studied silicones. Silicones used in palliative orthoses have higher instantaneous deformation-permanent strain and narrower creep-recovery distribution. The proposed methodology based on DMA, FDA and nonlinear regression is an useful tool to characterize and choose the proper silicone for each podiatry application according to their viscoelastic properties. Copyright © 2016 Elsevier Ltd. All rights reserved.

Prediction models for clustered data: comparison of a random intercept and standard regression model

PubMed Central

2013-01-01

Background When study data are clustered, standard regression analysis is considered inappropriate and analytical techniques for clustered data need to be used. For prediction research in which the interest of predictor effects is on the patient level, random effect regression models are probably preferred over standard regression analysis. It is well known that the random effect parameter estimates and the standard logistic regression parameter estimates are different. Here, we compared random effect and standard logistic regression models for their ability to provide accurate predictions. Methods Using an empirical study on 1642 surgical patients at risk of postoperative nausea and vomiting, who were treated by one of 19 anesthesiologists (clusters), we developed prognostic models either with standard or random intercept logistic regression. External validity of these models was assessed in new patients from other anesthesiologists. We supported our results with simulation studies using intra-class correlation coefficients (ICC) of 5%, 15%, or 30%. Standard performance measures and measures adapted for the clustered data structure were estimated. Results The model developed with random effect analysis showed better discrimination than the standard approach, if the cluster effects were used for risk prediction (standard c-index of 0.69 versus 0.66). In the external validation set, both models showed similar discrimination (standard c-index 0.68 versus 0.67). The simulation study confirmed these results. For datasets with a high ICC (≥15%), model calibration was only adequate in external subjects, if the used performance measure assumed the same data structure as the model development method: standard calibration measures showed good calibration for the standard developed model, calibration measures adapting the clustered data structure showed good calibration for the prediction model with random intercept. Conclusion The models with random intercept discriminate better than the standard model only if the cluster effect is used for predictions. The prediction model with random intercept had good calibration within clusters. PMID:23414436

Prediction models for clustered data: comparison of a random intercept and standard regression model.

PubMed

Bouwmeester, Walter; Twisk, Jos W R; Kappen, Teus H; van Klei, Wilton A; Moons, Karel G M; Vergouwe, Yvonne

2013-02-15

When study data are clustered, standard regression analysis is considered inappropriate and analytical techniques for clustered data need to be used. For prediction research in which the interest of predictor effects is on the patient level, random effect regression models are probably preferred over standard regression analysis. It is well known that the random effect parameter estimates and the standard logistic regression parameter estimates are different. Here, we compared random effect and standard logistic regression models for their ability to provide accurate predictions. Using an empirical study on 1642 surgical patients at risk of postoperative nausea and vomiting, who were treated by one of 19 anesthesiologists (clusters), we developed prognostic models either with standard or random intercept logistic regression. External validity of these models was assessed in new patients from other anesthesiologists. We supported our results with simulation studies using intra-class correlation coefficients (ICC) of 5%, 15%, or 30%. Standard performance measures and measures adapted for the clustered data structure were estimated. The model developed with random effect analysis showed better discrimination than the standard approach, if the cluster effects were used for risk prediction (standard c-index of 0.69 versus 0.66). In the external validation set, both models showed similar discrimination (standard c-index 0.68 versus 0.67). The simulation study confirmed these results. For datasets with a high ICC (≥15%), model calibration was only adequate in external subjects, if the used performance measure assumed the same data structure as the model development method: standard calibration measures showed good calibration for the standard developed model, calibration measures adapting the clustered data structure showed good calibration for the prediction model with random intercept. The models with random intercept discriminate better than the standard model only if the cluster effect is used for predictions. The prediction model with random intercept had good calibration within clusters.

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.

On the Asymptotic Relative Efficiency of Planned Missingness Designs.

PubMed

Rhemtulla, Mijke; Savalei, Victoria; Little, Todd D

2016-03-01

In planned missingness (PM) designs, certain data are set a priori to be missing. PM designs can increase validity and reduce cost; however, little is known about the loss of efficiency that accompanies these designs. The present paper compares PM designs to reduced sample (RN) designs that have the same total number of data points concentrated in fewer participants. In 4 studies, we consider models for both observed and latent variables, designs that do or do not include an "X set" of variables with complete data, and a full range of between- and within-set correlation values. All results are obtained using asymptotic relative efficiency formulas, and thus no data are generated; this novel approach allows us to examine whether PM designs have theoretical advantages over RN designs removing the impact of sampling error. Our primary findings are that (a) in manifest variable regression models, estimates of regression coefficients have much lower relative efficiency in PM designs as compared to RN designs, (b) relative efficiency of factor correlation or latent regression coefficient estimates is maximized when the indicators of each latent variable come from different sets, and (c) the addition of an X set improves efficiency in manifest variable regression models only for the parameters that directly involve the X-set variables, but it substantially improves efficiency of most parameters in latent variable models. We conclude that PM designs can be beneficial when the model of interest is a latent variable model; recommendations are made for how to optimize such a design.

Multilevel covariance regression with correlated random effects in the mean and variance structure.

PubMed

Quintero, Adrian; Lesaffre, Emmanuel

2017-09-01

Multivariate regression methods generally assume a constant covariance matrix for the observations. In case a heteroscedastic model is needed, the parametric and nonparametric covariance regression approaches can be restrictive in the literature. We propose a multilevel regression model for the mean and covariance structure, including random intercepts in both components and allowing for correlation between them. The implied conditional covariance function can be different across clusters as a result of the random effect in the variance structure. In addition, allowing for correlation between the random intercepts in the mean and covariance makes the model convenient for skewedly distributed responses. Furthermore, it permits us to analyse directly the relation between the mean response level and the variability in each cluster. Parameter estimation is carried out via Gibbs sampling. We compare the performance of our model to other covariance modelling approaches in a simulation study. Finally, the proposed model is applied to the RN4CAST dataset to identify the variables that impact burnout of nurses in Belgium. © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

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

Independent contrasts and PGLS regression estimators are equivalent.

PubMed

Blomberg, Simon P; Lefevre, James G; Wells, Jessie A; Waterhouse, Mary

2012-05-01

We prove that the slope parameter of the ordinary least squares regression of phylogenetically independent contrasts (PICs) conducted through the origin is identical to the slope parameter of the method of generalized least squares (GLSs) regression under a Brownian motion model of evolution. This equivalence has several implications: 1. Understanding the structure of the linear model for GLS regression provides insight into when and why phylogeny is important in comparative studies. 2. The limitations of the PIC regression analysis are the same as the limitations of the GLS model. In particular, phylogenetic covariance applies only to the response variable in the regression and the explanatory variable should be regarded as fixed. Calculation of PICs for explanatory variables should be treated as a mathematical idiosyncrasy of the PIC regression algorithm. 3. Since the GLS estimator is the best linear unbiased estimator (BLUE), the slope parameter estimated using PICs is also BLUE. 4. If the slope is estimated using different branch lengths for the explanatory and response variables in the PIC algorithm, the estimator is no longer the BLUE, so this is not recommended. Finally, we discuss whether or not and how to accommodate phylogenetic covariance in regression analyses, particularly in relation to the problem of phylogenetic uncertainty. This discussion is from both frequentist and Bayesian perspectives.

A matching framework to improve causal inference in interrupted time-series analysis.

PubMed

Linden, Ariel

2018-04-01

Interrupted time-series analysis (ITSA) is a popular evaluation methodology in which a single treatment unit's outcome is studied over time and the intervention is expected to "interrupt" the level and/or trend of the outcome, subsequent to its introduction. When ITSA is implemented without a comparison group, the internal validity may be quite poor. Therefore, adding a comparable control group to serve as the counterfactual is always preferred. This paper introduces a novel matching framework, ITSAMATCH, to create a comparable control group by matching directly on covariates and then use these matches in the outcomes model. We evaluate the effect of California's Proposition 99 (passed in 1988) for reducing cigarette sales, by comparing California to other states not exposed to smoking reduction initiatives. We compare ITSAMATCH results to 2 commonly used matching approaches, synthetic controls (SYNTH), and regression adjustment; SYNTH reweights nontreated units to make them comparable to the treated unit, and regression adjusts covariates directly. Methods are compared by assessing covariate balance and treatment effects. Both ITSAMATCH and SYNTH achieved covariate balance and estimated similar treatment effects. The regression model found no treatment effect and produced inconsistent covariate adjustment. While the matching framework achieved results comparable to SYNTH, it has the advantage of being technically less complicated, while producing statistical estimates that are straightforward to interpret. Conversely, regression adjustment may "adjust away" a treatment effect. Given its advantages, ITSAMATCH should be considered as a primary approach for evaluating treatment effects in multiple-group time-series analysis. © 2017 John Wiley & Sons, Ltd.

Multiple regression and Artificial Neural Network for long-term rainfall forecasting using large scale climate modes

NASA Astrophysics Data System (ADS)

Mekanik, F.; Imteaz, M. A.; Gato-Trinidad, S.; Elmahdi, A.

2013-10-01

In this study, the application of Artificial Neural Networks (ANN) and Multiple regression analysis (MR) to forecast long-term seasonal spring rainfall in Victoria, Australia was investigated using lagged El Nino Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD) as potential predictors. The use of dual (combined lagged ENSO-IOD) input sets for calibrating and validating ANN and MR Models is proposed to investigate the simultaneous effect of past values of these two major climate modes on long-term spring rainfall prediction. The MR models that did not violate the limits of statistical significance and multicollinearity were selected for future spring rainfall forecast. The ANN was developed in the form of multilayer perceptron using Levenberg-Marquardt algorithm. Both MR and ANN modelling were assessed statistically using mean square error (MSE), mean absolute error (MAE), Pearson correlation (r) and Willmott index of agreement (d). The developed MR and ANN models were tested on out-of-sample test sets; the MR models showed very poor generalisation ability for east Victoria with correlation coefficients of -0.99 to -0.90 compared to ANN with correlation coefficients of 0.42-0.93; ANN models also showed better generalisation ability for central and west Victoria with correlation coefficients of 0.68-0.85 and 0.58-0.97 respectively. The ability of multiple regression models to forecast out-of-sample sets is compatible with ANN for Daylesford in central Victoria and Kaniva in west Victoria (r = 0.92 and 0.67 respectively). The errors of the testing sets for ANN models are generally lower compared to multiple regression models. The statistical analysis suggest the potential of ANN over MR models for rainfall forecasting using large scale climate modes.

An EM-based semi-parametric mixture model approach to the regression analysis of competing-risks data.

PubMed

Ng, S K; McLachlan, G J

2003-04-15

We consider a mixture model approach to the regression analysis of competing-risks data. Attention is focused on inference concerning the effects of factors on both the probability of occurrence and the hazard rate conditional on each of the failure types. These two quantities are specified in the mixture model using the logistic model and the proportional hazards model, respectively. We propose a semi-parametric mixture method to estimate the logistic and regression coefficients jointly, whereby the component-baseline hazard functions are completely unspecified. Estimation is based on maximum likelihood on the basis of the full likelihood, implemented via an expectation-conditional maximization (ECM) algorithm. Simulation studies are performed to compare the performance of the proposed semi-parametric method with a fully parametric mixture approach. The results show that when the component-baseline hazard is monotonic increasing, the semi-parametric and fully parametric mixture approaches are comparable for mildly and moderately censored samples. When the component-baseline hazard is not monotonic increasing, the semi-parametric method consistently provides less biased estimates than a fully parametric approach and is comparable in efficiency in the estimation of the parameters for all levels of censoring. The methods are illustrated using a real data set of prostate cancer patients treated with different dosages of the drug diethylstilbestrol. Copyright 2003 John Wiley & Sons, Ltd.

Application of seemingly unrelated regression in medical data with intermittently observed time-dependent covariates.

PubMed

Keshavarzi, Sareh; Ayatollahi, Seyyed Mohammad Taghi; Zare, Najaf; Pakfetrat, Maryam

2012-01-01

BACKGROUND. In many studies with longitudinal data, time-dependent covariates can only be measured intermittently (not at all observation times), and this presents difficulties for standard statistical analyses. This situation is common in medical studies, and methods that deal with this challenge would be useful. METHODS. In this study, we performed the seemingly unrelated regression (SUR) based models, with respect to each observation time in longitudinal data with intermittently observed time-dependent covariates and further compared these models with mixed-effect regression models (MRMs) under three classic imputation procedures. Simulation studies were performed to compare the sample size properties of the estimated coefficients for different modeling choices. RESULTS. In general, the proposed models in the presence of intermittently observed time-dependent covariates showed a good performance. However, when we considered only the observed values of the covariate without any imputations, the resulted biases were greater. The performances of the proposed SUR-based models in comparison with MRM using classic imputation methods were nearly similar with approximately equal amounts of bias and MSE. CONCLUSION. The simulation study suggests that the SUR-based models work as efficiently as MRM in the case of intermittently observed time-dependent covariates. Thus, it can be used as an alternative to MRM.

Coronary artery calcium distributions in older persons in the AGES-Reykjavik study

PubMed Central

Gudmundsson, Elias Freyr; Gudnason, Vilmundur; Sigurdsson, Sigurdur; Launer, Lenore J.; Harris, Tamara B.; Aspelund, Thor

2013-01-01

Coronary Artery Calcium (CAC) is a sign of advanced atherosclerosis and an independent risk factor for cardiac events. Here, we describe CAC-distributions in an unselected aged population and compare modelling methods to characterize CAC-distribution. CAC is difficult to model because it has a skewed and zero inflated distribution with over-dispersion. Data are from the AGES-Reykjavik sample, a large population based study [2002-2006] in Iceland of 5,764 persons aged 66-96 years. Linear regressions using logarithmic- and Box-Cox transformations on CAC+1, quantile regression and a Zero-Inflated Negative Binomial model (ZINB) were applied. Methods were compared visually and with the PRESS-statistic, R2 and number of detected associations with concurrently measured variables. There were pronounced differences in CAC according to sex, age, history of coronary events and presence of plaque in the carotid artery. Associations with conventional coronary artery disease (CAD) risk factors varied between the sexes. The ZINB model provided the best results with respect to the PRESS-statistic, R2, and predicted proportion of zero scores. The ZINB model detected similar numbers of associations as the linear regression on ln(CAC+1) and usually with the same risk factors. PMID:22990371

A support vector regression-firefly algorithm-based model for limiting velocity prediction in sewer pipes.

PubMed

Ebtehaj, Isa; Bonakdari, Hossein

2016-01-01

Sediment transport without deposition is an essential consideration in the optimum design of sewer pipes. In this study, a novel method based on a combination of support vector regression (SVR) and the firefly algorithm (FFA) is proposed to predict the minimum velocity required to avoid sediment settling in pipe channels, which is expressed as the densimetric Froude number (Fr). The efficiency of support vector machine (SVM) models depends on the suitable selection of SVM parameters. In this particular study, FFA is used by determining these SVM parameters. The actual effective parameters on Fr calculation are generally identified by employing dimensional analysis. The different dimensionless variables along with the models are introduced. The best performance is attributed to the model that employs the sediment volumetric concentration (C(V)), ratio of relative median diameter of particles to hydraulic radius (d/R), dimensionless particle number (D(gr)) and overall sediment friction factor (λ(s)) parameters to estimate Fr. The performance of the SVR-FFA model is compared with genetic programming, artificial neural network and existing regression-based equations. The results indicate the superior performance of SVR-FFA (mean absolute percentage error = 2.123%; root mean square error =0.116) compared with other methods.

Collapse susceptibility mapping in karstified gypsum terrain (Sivas basin - Turkey) by conditional probability, logistic regression, artificial neural network models

NASA Astrophysics Data System (ADS)

Yilmaz, Isik; Keskin, Inan; Marschalko, Marian; Bednarik, Martin

2010-05-01

This study compares the GIS based collapse susceptibility mapping methods such as; conditional probability (CP), logistic regression (LR) and artificial neural networks (ANN) applied in gypsum rock masses in Sivas basin (Turkey). Digital Elevation Model (DEM) was first constructed using GIS software. Collapse-related factors, directly or indirectly related to the causes of collapse occurrence, such as distance from faults, slope angle and aspect, topographical elevation, distance from drainage, topographic wetness index- TWI, stream power index- SPI, Normalized Difference Vegetation Index (NDVI) by means of vegetation cover, distance from roads and settlements were used in the collapse susceptibility analyses. In the last stage of the analyses, collapse susceptibility maps were produced from CP, LR and ANN models, and they were then compared by means of their validations. Area Under Curve (AUC) values obtained from all three methodologies showed that the map obtained from ANN model looks like more accurate than the other models, and the results also showed that the artificial neural networks is a usefull tool in preparation of collapse susceptibility map and highly compatible with GIS operating features. Key words: Collapse; doline; susceptibility map; gypsum; GIS; conditional probability; logistic regression; artificial neural networks.

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.

A phenomenological biological dose model for proton therapy based on linear energy transfer spectra.

PubMed

Rørvik, Eivind; Thörnqvist, Sara; Stokkevåg, Camilla H; Dahle, Tordis J; Fjaera, Lars Fredrik; Ytre-Hauge, Kristian S

2017-06-01

The relative biological effectiveness (RBE) of protons varies with the radiation quality, quantified by the linear energy transfer (LET). Most phenomenological models employ a linear dependency of the dose-averaged LET (LET d ) to calculate the biological dose. However, several experiments have indicated a possible non-linear trend. Our aim was to investigate if biological dose models including non-linear LET dependencies should be considered, by introducing a LET spectrum based dose model. The RBE-LET relationship was investigated by fitting of polynomials from 1st to 5th degree to a database of 85 data points from aerobic in vitro experiments. We included both unweighted and weighted regression, the latter taking into account experimental uncertainties. Statistical testing was performed to decide whether higher degree polynomials provided better fits to the data as compared to lower degrees. The newly developed models were compared to three published LET d based models for a simulated spread out Bragg peak (SOBP) scenario. The statistical analysis of the weighted regression analysis favored a non-linear RBE-LET relationship, with the quartic polynomial found to best represent the experimental data (P = 0.010). The results of the unweighted regression analysis were on the borderline of statistical significance for non-linear functions (P = 0.053), and with the current database a linear dependency could not be rejected. For the SOBP scenario, the weighted non-linear model estimated a similar mean RBE value (1.14) compared to the three established models (1.13-1.17). The unweighted model calculated a considerably higher RBE value (1.22). The analysis indicated that non-linear models could give a better representation of the RBE-LET relationship. However, this is not decisive, as inclusion of the experimental uncertainties in the regression analysis had a significant impact on the determination and ranking of the models. As differences between the models were observed for the SOBP scenario, both non-linear LET spectrum- and linear LET d based models should be further evaluated in clinically realistic scenarios. © 2017 American Association of Physicists in Medicine.

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.

Using Bayesian regression to test hypotheses about relationships between parameters and covariates in cognitive models.

PubMed

Boehm, Udo; Steingroever, Helen; Wagenmakers, Eric-Jan

2018-06-01

An important tool in the advancement of cognitive science are quantitative models that represent different cognitive variables in terms of model parameters. To evaluate such models, their parameters are typically tested for relationships with behavioral and physiological variables that are thought to reflect specific cognitive processes. However, many models do not come equipped with the statistical framework needed to relate model parameters to covariates. Instead, researchers often revert to classifying participants into groups depending on their values on the covariates, and subsequently comparing the estimated model parameters between these groups. Here we develop a comprehensive solution to the covariate problem in the form of a Bayesian regression framework. Our framework can be easily added to existing cognitive models and allows researchers to quantify the evidential support for relationships between covariates and model parameters using Bayes factors. Moreover, we present a simulation study that demonstrates the superiority of the Bayesian regression framework to the conventional classification-based approach.

Application of Bayesian methods to habitat selection modeling of the northern spotted owl in California: new statistical methods for wildlife research

Treesearch

Howard B. Stauffer; Cynthia J. Zabel; Jeffrey R. Dunk

2005-01-01

We compared a set of competing logistic regression habitat selection models for Northern Spotted Owls (Strix occidentalis caurina) in California. The habitat selection models were estimated, compared, evaluated, and tested using multiple sample datasets collected on federal forestlands in northern California. We used Bayesian methods in interpreting...

Adjusting for Confounding in Early Postlaunch Settings: Going Beyond Logistic Regression Models.

PubMed

Schmidt, Amand F; Klungel, Olaf H; Groenwold, Rolf H H

2016-01-01

Postlaunch data on medical treatments can be analyzed to explore adverse events or relative effectiveness in real-life settings. These analyses are often complicated by the number of potential confounders and the possibility of model misspecification. We conducted a simulation study to compare the performance of logistic regression, propensity score, disease risk score, and stabilized inverse probability weighting methods to adjust for confounding. Model misspecification was induced in the independent derivation dataset. We evaluated performance using relative bias confidence interval coverage of the true effect, among other metrics. At low events per coefficient (1.0 and 0.5), the logistic regression estimates had a large relative bias (greater than -100%). Bias of the disease risk score estimates was at most 13.48% and 18.83%. For the propensity score model, this was 8.74% and >100%, respectively. At events per coefficient of 1.0 and 0.5, inverse probability weighting frequently failed or reduced to a crude regression, resulting in biases of -8.49% and 24.55%. Coverage of logistic regression estimates became less than the nominal level at events per coefficient ≤5. For the disease risk score, inverse probability weighting, and propensity score, coverage became less than nominal at events per coefficient ≤2.5, ≤1.0, and ≤1.0, respectively. Bias of misspecified disease risk score models was 16.55%. In settings with low events/exposed subjects per coefficient, disease risk score methods can be useful alternatives to logistic regression models, especially when propensity score models cannot be used. Despite better performance of disease risk score methods than logistic regression and propensity score models in small events per coefficient settings, bias, and coverage still deviated from nominal.

An empirical study of statistical properties of variance partition coefficients for multi-level logistic regression models

USGS Publications Warehouse

Li, Ji; Gray, B.R.; Bates, D.M.

2008-01-01

Partitioning the variance of a response by design levels is challenging for binomial and other discrete outcomes. Goldstein (2003) proposed four definitions for variance partitioning coefficients (VPC) under a two-level logistic regression model. In this study, we explicitly derived formulae for multi-level logistic regression model and subsequently studied the distributional properties of the calculated VPCs. Using simulations and a vegetation dataset, we demonstrated associations between different VPC definitions, the importance of methods for estimating VPCs (by comparing VPC obtained using Laplace and penalized quasilikehood methods), and bivariate dependence between VPCs calculated at different levels. Such an empirical study lends an immediate support to wider applications of VPC in scientific data analysis.

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.

Replica analysis of overfitting in regression models for time-to-event data

NASA Astrophysics Data System (ADS)

Coolen, A. C. C.; Barrett, J. E.; Paga, P.; Perez-Vicente, C. J.

2017-09-01

Overfitting, which happens when the number of parameters in a model is too large compared to the number of data points available for determining these parameters, is a serious and growing problem in survival analysis. While modern medicine presents us with data of unprecedented dimensionality, these data cannot yet be used effectively for clinical outcome prediction. Standard error measures in maximum likelihood regression, such as p-values and z-scores, are blind to overfitting, and even for Cox’s proportional hazards model (the main tool of medical statisticians), one finds in literature only rules of thumb on the number of samples required to avoid overfitting. In this paper we present a mathematical theory of overfitting in regression models for time-to-event data, which aims to increase our quantitative understanding of the problem and provide practical tools with which to correct regression outcomes for the impact of overfitting. It is based on the replica method, a statistical mechanical technique for the analysis of heterogeneous many-variable systems that has been used successfully for several decades in physics, biology, and computer science, but not yet in medical statistics. We develop the theory initially for arbitrary regression models for time-to-event data, and verify its predictions in detail for the popular Cox model.

Boosting structured additive quantile regression for longitudinal childhood obesity data.

PubMed

Fenske, Nora; Fahrmeir, Ludwig; Hothorn, Torsten; Rzehak, Peter; Höhle, Michael

2013-07-25

Childhood obesity and the investigation of its risk factors has become an important public health issue. Our work is based on and motivated by a German longitudinal study including 2,226 children with up to ten measurements on their body mass index (BMI) and risk factors from birth to the age of 10 years. We introduce boosting of structured additive quantile regression as a novel distribution-free approach for longitudinal quantile regression. The quantile-specific predictors of our model include conventional linear population effects, smooth nonlinear functional effects, varying-coefficient terms, and individual-specific effects, such as intercepts and slopes. Estimation is based on boosting, a computer intensive inference method for highly complex models. We propose a component-wise functional gradient descent boosting algorithm that allows for penalized estimation of the large variety of different effects, particularly leading to individual-specific effects shrunken toward zero. This concept allows us to flexibly estimate the nonlinear age curves of upper quantiles of the BMI distribution, both on population and on individual-specific level, adjusted for further risk factors and to detect age-varying effects of categorical risk factors. Our model approach can be regarded as the quantile regression analog of Gaussian additive mixed models (or structured additive mean regression models), and we compare both model classes with respect to our obesity data.

Comprehensive ripeness-index for prediction of ripening level in mangoes by multivariate modelling of ripening behaviour

NASA Astrophysics Data System (ADS)

Eyarkai Nambi, Vijayaram; Thangavel, Kuladaisamy; Manickavasagan, Annamalai; Shahir, Sultan

2017-01-01

Prediction of ripeness level in climacteric fruits is essential for post-harvest handling. An index capable of predicting ripening level with minimum inputs would be highly beneficial to the handlers, processors and researchers in fruit industry. A study was conducted with Indian mango cultivars to develop a ripeness index and associated model. Changes in physicochemical, colour and textural properties were measured throughout the ripening period and the period was classified into five stages (unripe, early ripe, partially ripe, ripe and over ripe). Multivariate regression techniques like partial least square regression, principal component regression and multi linear regression were compared and evaluated for its prediction. Multi linear regression model with 12 parameters was found more suitable in ripening prediction. Scientific variable reduction method was adopted to simplify the developed model. Better prediction was achieved with either 2 or 3 variables (total soluble solids, colour and acidity). Cross validation was done to increase the robustness and it was found that proposed ripening index was more effective in prediction of ripening stages. Three-variable model would be suitable for commercial applications where reasonable accuracies are sufficient. However, 12-variable model can be used to obtain more precise results in research and development applications.

A Powerful Test for Comparing Multiple Regression Functions.

PubMed

Maity, Arnab

2012-09-01

In this article, we address the important problem of comparison of two or more population regression functions. Recently, Pardo-Fernández, Van Keilegom and González-Manteiga (2007) developed test statistics for simple nonparametric regression models: Y(ij) = θ(j)(Z(ij)) + σ(j)(Z(ij))∊(ij), based on empirical distributions of the errors in each population j = 1, … , J. In this paper, we propose a test for equality of the θ(j)(·) based on the concept of generalized likelihood ratio type statistics. We also generalize our test for other nonparametric regression setups, e.g, nonparametric logistic regression, where the loglikelihood for population j is any general smooth function [Formula: see text]. We describe a resampling procedure to obtain the critical values of the test. In addition, we present a simulation study to evaluate the performance of the proposed test and compare our results to those in Pardo-Fernández et al. (2007).

Prediction of monthly rainfall in Victoria, Australia: Clusterwise linear regression approach

NASA Astrophysics Data System (ADS)

Bagirov, Adil M.; Mahmood, Arshad; Barton, Andrew

2017-05-01

This paper develops the Clusterwise Linear Regression (CLR) technique for prediction of monthly rainfall. The CLR is a combination of clustering and regression techniques. It is formulated as an optimization problem and an incremental algorithm is designed to solve it. The algorithm is applied to predict monthly rainfall in Victoria, Australia using rainfall data with five input meteorological variables over the period of 1889-2014 from eight geographically diverse weather stations. The prediction performance of the CLR method is evaluated by comparing observed and predicted rainfall values using four measures of forecast accuracy. The proposed method is also compared with the CLR using the maximum likelihood framework by the expectation-maximization algorithm, multiple linear regression, artificial neural networks and the support vector machines for regression models using computational results. The results demonstrate that the proposed algorithm outperforms other methods in most locations.

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.

Selecting risk factors: a comparison of discriminant analysis, logistic regression and Cox's regression model using data from the Tromsø Heart Study.

PubMed

Brenn, T; Arnesen, E

1985-01-01

For comparative evaluation, discriminant analysis, logistic regression and Cox's model were used to select risk factors for total and coronary deaths among 6595 men aged 20-49 followed for 9 years. Groups with mortality between 5 and 93 per 1000 were considered. Discriminant analysis selected variable sets only marginally different from the logistic and Cox methods which always selected the same sets. A time-saving option, offered for both the logistic and Cox selection, showed no advantage compared with discriminant analysis. Analysing more than 3800 subjects, the logistic and Cox methods consumed, respectively, 80 and 10 times more computer time than discriminant analysis. When including the same set of variables in non-stepwise analyses, all methods estimated coefficients that in most cases were almost identical. In conclusion, discriminant analysis is advocated for preliminary or stepwise analysis, otherwise Cox's method should be used.

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.

Estimation of sex and stature using anthropometry of the upper extremity in an Australian population.

PubMed

Howley, Donna; Howley, Peter; Oxenham, Marc F

2018-06-01

Stature and a further 8 anthropometric dimensions were recorded from the arms and hands of a sample of 96 staff and students from the Australian National University and The University of Newcastle, Australia. These dimensions were used to create simple and multiple logistic regression models for sex estimation and simple and multiple linear regression equations for stature estimation of a contemporary Australian population. Overall sex classification accuracies using the models created were comparable to similar studies. The stature estimation models achieved standard errors of estimates (SEE) which were comparable to and in many cases lower than those achieved in similar research. Generic, non sex-specific models achieved similar SEEs and R 2 values to the sex-specific models indicating stature may be accurately estimated when sex is unknown. Copyright © 2018 Elsevier B.V. All rights reserved.

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.

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.

Modeling Fire Occurrence at the City Scale: A Comparison between Geographically Weighted Regression and Global Linear Regression.

PubMed

Song, Chao; Kwan, Mei-Po; Zhu, Jiping

2017-04-08

An increasing number of fires are occurring with the rapid development of cities, resulting in increased risk for human beings and the environment. This study compares geographically weighted regression-based models, including geographically weighted regression (GWR) and geographically and temporally weighted regression (GTWR), which integrates spatial and temporal effects and global linear regression models (LM) for modeling fire risk at the city scale. The results show that the road density and the spatial distribution of enterprises have the strongest influences on fire risk, which implies that we should focus on areas where roads and enterprises are densely clustered. In addition, locations with a large number of enterprises have fewer fire ignition records, probably because of strict management and prevention measures. A changing number of significant variables across space indicate that heterogeneity mainly exists in the northern and eastern rural and suburban areas of Hefei city, where human-related facilities or road construction are only clustered in the city sub-centers. GTWR can capture small changes in the spatiotemporal heterogeneity of the variables while GWR and LM cannot. An approach that integrates space and time enables us to better understand the dynamic changes in fire risk. Thus governments can use the results to manage fire safety at the city scale.

Modeling Fire Occurrence at the City Scale: A Comparison between Geographically Weighted Regression and Global Linear Regression

PubMed Central

Song, Chao; Kwan, Mei-Po; Zhu, Jiping

2017-01-01

An increasing number of fires are occurring with the rapid development of cities, resulting in increased risk for human beings and the environment. This study compares geographically weighted regression-based models, including geographically weighted regression (GWR) and geographically and temporally weighted regression (GTWR), which integrates spatial and temporal effects and global linear regression models (LM) for modeling fire risk at the city scale. The results show that the road density and the spatial distribution of enterprises have the strongest influences on fire risk, which implies that we should focus on areas where roads and enterprises are densely clustered. In addition, locations with a large number of enterprises have fewer fire ignition records, probably because of strict management and prevention measures. A changing number of significant variables across space indicate that heterogeneity mainly exists in the northern and eastern rural and suburban areas of Hefei city, where human-related facilities or road construction are only clustered in the city sub-centers. GTWR can capture small changes in the spatiotemporal heterogeneity of the variables while GWR and LM cannot. An approach that integrates space and time enables us to better understand the dynamic changes in fire risk. Thus governments can use the results to manage fire safety at the city scale. PMID:28397745

Empirical Likelihood in Nonignorable Covariate-Missing Data Problems.

PubMed

Xie, Yanmei; Zhang, Biao

2017-04-20

Missing covariate data occurs often in regression analysis, which frequently arises in the health and social sciences as well as in survey sampling. We study methods for the analysis of a nonignorable covariate-missing data problem in an assumed conditional mean function when some covariates are completely observed but other covariates are missing for some subjects. We adopt the semiparametric perspective of Bartlett et al. (Improving upon the efficiency of complete case analysis when covariates are MNAR. Biostatistics 2014;15:719-30) on regression analyses with nonignorable missing covariates, in which they have introduced the use of two working models, the working probability model of missingness and the working conditional score model. In this paper, we study an empirical likelihood approach to nonignorable covariate-missing data problems with the objective of effectively utilizing the two working models in the analysis of covariate-missing data. We propose a unified approach to constructing a system of unbiased estimating equations, where there are more equations than unknown parameters of interest. One useful feature of these unbiased estimating equations is that they naturally incorporate the incomplete data into the data analysis, making it possible to seek efficient estimation of the parameter of interest even when the working regression function is not specified to be the optimal regression function. We apply the general methodology of empirical likelihood to optimally combine these unbiased estimating equations. We propose three maximum empirical likelihood estimators of the underlying regression parameters and compare their efficiencies with other existing competitors. We present a simulation study to compare the finite-sample performance of various methods with respect to bias, efficiency, and robustness to model misspecification. The proposed empirical likelihood method is also illustrated by an analysis of a data set from the US National Health and Nutrition Examination Survey (NHANES).

Factor analysis and multiple regression between topography and precipitation on Jeju Island, Korea

NASA Astrophysics Data System (ADS)

Um, Myoung-Jin; Yun, Hyeseon; Jeong, Chang-Sam; Heo, Jun-Haeng

2011-11-01

SummaryIn this study, new factors that influence precipitation were extracted from geographic variables using factor analysis, which allow for an accurate estimation of orographic precipitation. Correlation analysis was also used to examine the relationship between nine topographic variables from digital elevation models (DEMs) and the precipitation in Jeju Island. In addition, a spatial analysis was performed in order to verify the validity of the regression model. From the results of the correlation analysis, it was found that all of the topographic variables had a positive correlation with the precipitation. The relations between the variables also changed in accordance with a change in the precipitation duration. However, upon examining the correlation matrix, no significant relationship between the latitude and the aspect was found. According to the factor analysis, eight topographic variables (latitude being the exception) were found to have a direct influence on the precipitation. Three factors were then extracted from the eight topographic variables. By directly comparing the multiple regression model with the factors (model 1) to the multiple regression model with the topographic variables (model 3), it was found that model 1 did not violate the limits of statistical significance and multicollinearity. As such, model 1 was considered to be appropriate for estimating the precipitation when taking into account the topography. In the study of model 1, the multiple regression model using factor analysis was found to be the best method for estimating the orographic precipitation on Jeju Island.

Functional Data Analysis Applied to Modeling of Severe Acute Mucositis and Dysphagia Resulting From Head and Neck Radiation Therapy

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

Dean, Jamie A., E-mail: jamie.dean@icr.ac.uk; Wong, Kee H.; Gay, Hiram

Purpose: Current normal tissue complication probability modeling using logistic regression suffers from bias and high uncertainty in the presence of highly correlated radiation therapy (RT) dose data. This hinders robust estimates of dose-response associations and, hence, optimal normal tissue–sparing strategies from being elucidated. Using functional data analysis (FDA) to reduce the dimensionality of the dose data could overcome this limitation. Methods and Materials: FDA was applied to modeling of severe acute mucositis and dysphagia resulting from head and neck RT. Functional partial least squares regression (FPLS) and functional principal component analysis were used for dimensionality reduction of the dose-volume histogrammore » data. The reduced dose data were input into functional logistic regression models (functional partial least squares–logistic regression [FPLS-LR] and functional principal component–logistic regression [FPC-LR]) along with clinical data. This approach was compared with penalized logistic regression (PLR) in terms of predictive performance and the significance of treatment covariate–response associations, assessed using bootstrapping. Results: The area under the receiver operating characteristic curve for the PLR, FPC-LR, and FPLS-LR models was 0.65, 0.69, and 0.67, respectively, for mucositis (internal validation) and 0.81, 0.83, and 0.83, respectively, for dysphagia (external validation). The calibration slopes/intercepts for the PLR, FPC-LR, and FPLS-LR models were 1.6/−0.67, 0.45/0.47, and 0.40/0.49, respectively, for mucositis (internal validation) and 2.5/−0.96, 0.79/−0.04, and 0.79/0.00, respectively, for dysphagia (external validation). The bootstrapped odds ratios indicated significant associations between RT dose and severe toxicity in the mucositis and dysphagia FDA models. Cisplatin was significantly associated with severe dysphagia in the FDA models. None of the covariates was significantly associated with severe toxicity in the PLR models. Dose levels greater than approximately 1.0 Gy/fraction were most strongly associated with severe acute mucositis and dysphagia in the FDA models. Conclusions: FPLS and functional principal component analysis marginally improved predictive performance compared with PLR and provided robust dose-response associations. FDA is recommended for use in normal tissue complication probability modeling.« less

Functional Data Analysis Applied to Modeling of Severe Acute Mucositis and Dysphagia Resulting From Head and Neck Radiation Therapy.

PubMed

Dean, Jamie A; Wong, Kee H; Gay, Hiram; Welsh, Liam C; Jones, Ann-Britt; Schick, Ulrike; Oh, Jung Hun; Apte, Aditya; Newbold, Kate L; Bhide, Shreerang A; Harrington, Kevin J; Deasy, Joseph O; Nutting, Christopher M; Gulliford, Sarah L

2016-11-15

Current normal tissue complication probability modeling using logistic regression suffers from bias and high uncertainty in the presence of highly correlated radiation therapy (RT) dose data. This hinders robust estimates of dose-response associations and, hence, optimal normal tissue-sparing strategies from being elucidated. Using functional data analysis (FDA) to reduce the dimensionality of the dose data could overcome this limitation. FDA was applied to modeling of severe acute mucositis and dysphagia resulting from head and neck RT. Functional partial least squares regression (FPLS) and functional principal component analysis were used for dimensionality reduction of the dose-volume histogram data. The reduced dose data were input into functional logistic regression models (functional partial least squares-logistic regression [FPLS-LR] and functional principal component-logistic regression [FPC-LR]) along with clinical data. This approach was compared with penalized logistic regression (PLR) in terms of predictive performance and the significance of treatment covariate-response associations, assessed using bootstrapping. The area under the receiver operating characteristic curve for the PLR, FPC-LR, and FPLS-LR models was 0.65, 0.69, and 0.67, respectively, for mucositis (internal validation) and 0.81, 0.83, and 0.83, respectively, for dysphagia (external validation). The calibration slopes/intercepts for the PLR, FPC-LR, and FPLS-LR models were 1.6/-0.67, 0.45/0.47, and 0.40/0.49, respectively, for mucositis (internal validation) and 2.5/-0.96, 0.79/-0.04, and 0.79/0.00, respectively, for dysphagia (external validation). The bootstrapped odds ratios indicated significant associations between RT dose and severe toxicity in the mucositis and dysphagia FDA models. Cisplatin was significantly associated with severe dysphagia in the FDA models. None of the covariates was significantly associated with severe toxicity in the PLR models. Dose levels greater than approximately 1.0 Gy/fraction were most strongly associated with severe acute mucositis and dysphagia in the FDA models. FPLS and functional principal component analysis marginally improved predictive performance compared with PLR and provided robust dose-response associations. FDA is recommended for use in normal tissue complication probability modeling. Copyright © 2016 The Author(s). Published by Elsevier Inc. All rights reserved.

Spatiotemporal variability of urban growth factors: A global and local perspective on the megacity of Mumbai

NASA Astrophysics Data System (ADS)

Shafizadeh-Moghadam, Hossein; Helbich, Marco

2015-03-01

The rapid growth of megacities requires special attention among urban planners worldwide, and particularly in Mumbai, India, where growth is very pronounced. To cope with the planning challenges this will bring, developing a retrospective understanding of urban land-use dynamics and the underlying driving-forces behind urban growth is a key prerequisite. This research uses regression-based land-use change models - and in particular non-spatial logistic regression models (LR) and auto-logistic regression models (ALR) - for the Mumbai region over the period 1973-2010, in order to determine the drivers behind spatiotemporal urban expansion. Both global models are complemented by a local, spatial model, the so-called geographically weighted logistic regression (GWLR) model, one that explicitly permits variations in driving-forces across space. The study comes to two main conclusions. First, both global models suggest similar driving-forces behind urban growth over time, revealing that LRs and ALRs result in estimated coefficients with comparable magnitudes. Second, all the local coefficients show distinctive temporal and spatial variations. It is therefore concluded that GWLR aids our understanding of urban growth processes, and so can assist context-related planning and policymaking activities when seeking to secure a sustainable urban future.

Markovian prediction of future values for food grains in the economic survey

NASA Astrophysics Data System (ADS)

Sathish, S.; Khadar Babu, S. K.

2017-11-01

Now-a-days prediction and forecasting are plays a vital role in research. For prediction, regression is useful to predict the future value and current value on production process. In this paper, we assume food grain production exhibit Markov chain dependency and time homogeneity. The economic generative performance evaluation the balance time artificial fertilization different level in Estrusdetection using a daily Markov chain model. Finally, Markov process prediction gives better performance compare with Regression model.

Quantile regression for the statistical analysis of immunological data with many non-detects.

PubMed

Eilers, Paul H C; Röder, Esther; Savelkoul, Huub F J; van Wijk, Roy Gerth

2012-07-07

Immunological parameters are hard to measure. A well-known problem is the occurrence of values below the detection limit, the non-detects. Non-detects are a nuisance, because classical statistical analyses, like ANOVA and regression, cannot be applied. The more advanced statistical techniques currently available for the analysis of datasets with non-detects can only be used if a small percentage of the data are non-detects. Quantile regression, a generalization of percentiles to regression models, models the median or higher percentiles and tolerates very high numbers of non-detects. We present a non-technical introduction and illustrate it with an implementation to real data from a clinical trial. We show that by using quantile regression, groups can be compared and that meaningful linear trends can be computed, even if more than half of the data consists of non-detects. Quantile regression is a valuable addition to the statistical methods that can be used for the analysis of immunological datasets with non-detects.

Does Bootstrap Procedure Provide Biased Estimates? An Empirical Examination for a Case of Multiple Regression.

ERIC Educational Resources Information Center

Fan, Xitao

This paper empirically and systematically assessed the performance of bootstrap resampling procedure as it was applied to a regression model. Parameter estimates from Monte Carlo experiments (repeated sampling from population) and bootstrap experiments (repeated resampling from one original bootstrap sample) were generated and compared. Sample…

Goodness of Fit and Misspecification in Quantile Regressions

ERIC Educational Resources Information Center

Furno, Marilena

2011-01-01

The article considers a test of specification for quantile regressions. The test relies on the increase of the objective function and the worsening of the fit when unnecessary constraints are imposed. It compares the objective functions of restricted and unrestricted models and, in its different formulations, it verifies (a) forecast ability, (b)…

The novel application of artificial neural network on bioelectrical impedance analysis to assess the body composition in elderly

PubMed Central

2013-01-01

Background This study aims to improve accuracy of Bioelectrical Impedance Analysis (BIA) prediction equations for estimating fat free mass (FFM) of the elderly by using non-linear Back Propagation Artificial Neural Network (BP-ANN) model and to compare the predictive accuracy with the linear regression model by using energy dual X-ray absorptiometry (DXA) as reference method. Methods A total of 88 Taiwanese elderly adults were recruited in this study as subjects. Linear regression equations and BP-ANN prediction equation were developed using impedances and other anthropometrics for predicting the reference FFM measured by DXA (FFMDXA) in 36 male and 26 female Taiwanese elderly adults. The FFM estimated by BIA prediction equations using traditional linear regression model (FFMLR) and BP-ANN model (FFMANN) were compared to the FFMDXA. The measuring results of an additional 26 elderly adults were used to validate than accuracy of the predictive models. Results The results showed the significant predictors were impedance, gender, age, height and weight in developed FFMLR linear model (LR) for predicting FFM (coefficient of determination, r2 = 0.940; standard error of estimate (SEE) = 2.729 kg; root mean square error (RMSE) = 2.571kg, P < 0.001). The above predictors were set as the variables of the input layer by using five neurons in the BP-ANN model (r2 = 0.987 with a SD = 1.192 kg and relatively lower RMSE = 1.183 kg), which had greater (improved) accuracy for estimating FFM when compared with linear model. The results showed a better agreement existed between FFMANN and FFMDXA than that between FFMLR and FFMDXA. Conclusion When compared the performance of developed prediction equations for estimating reference FFMDXA, the linear model has lower r2 with a larger SD in predictive results than that of BP-ANN model, which indicated ANN model is more suitable for estimating FFM. PMID:23388042

Method selection and adaptation for distributed monitoring of infectious diseases for syndromic surveillance.

PubMed

Xing, Jian; Burkom, Howard; Tokars, Jerome

2011-12-01

Automated surveillance systems require statistical methods to recognize increases in visit counts that might indicate an outbreak. In prior work we presented methods to enhance the sensitivity of C2, a commonly used time series method. In this study, we compared the enhanced C2 method with five regression models. We used emergency department chief complaint data from US CDC BioSense surveillance system, aggregated by city (total of 206 hospitals, 16 cities) during 5/2008-4/2009. Data for six syndromes (asthma, gastrointestinal, nausea and vomiting, rash, respiratory, and influenza-like illness) was used and was stratified by mean count (1-19, 20-49, ≥50 per day) into 14 syndrome-count categories. We compared the sensitivity for detecting single-day artificially-added increases in syndrome counts. Four modifications of the C2 time series method, and five regression models (two linear and three Poisson), were tested. A constant alert rate of 1% was used for all methods. Among the regression models tested, we found that a Poisson model controlling for the logarithm of total visits (i.e., visits both meeting and not meeting a syndrome definition), day of week, and 14-day time period was best. Among 14 syndrome-count categories, time series and regression methods produced approximately the same sensitivity (<5% difference) in 6; in six categories, the regression method had higher sensitivity (range 6-14% improvement), and in two categories the time series method had higher sensitivity. When automated data are aggregated to the city level, a Poisson regression model that controls for total visits produces the best overall sensitivity for detecting artificially added visit counts. This improvement was achieved without increasing the alert rate, which was held constant at 1% for all methods. These findings will improve our ability to detect outbreaks in automated surveillance system data. Published by Elsevier Inc.

Classification and regression tree analysis of acute-on-chronic hepatitis B liver failure: Seeing the forest for the trees.

PubMed

Shi, K-Q; Zhou, Y-Y; Yan, H-D; Li, H; Wu, F-L; Xie, Y-Y; Braddock, M; Lin, X-Y; Zheng, M-H

2017-02-01

At present, there is no ideal model for predicting the short-term outcome of patients with acute-on-chronic hepatitis B liver failure (ACHBLF). This study aimed to establish and validate a prognostic model by using the classification and regression tree (CART) analysis. A total of 1047 patients from two separate medical centres with suspected ACHBLF were screened in the study, which were recognized as derivation cohort and validation cohort, respectively. CART analysis was applied to predict the 3-month mortality of patients with ACHBLF. The accuracy of the CART model was tested using the area under the receiver operating characteristic curve, which was compared with the model for end-stage liver disease (MELD) score and a new logistic regression model. CART analysis identified four variables as prognostic factors of ACHBLF: total bilirubin, age, serum sodium and INR, and three distinct risk groups: low risk (4.2%), intermediate risk (30.2%-53.2%) and high risk (81.4%-96.9%). The new logistic regression model was constructed with four independent factors, including age, total bilirubin, serum sodium and prothrombin activity by multivariate logistic regression analysis. The performances of the CART model (0.896), similar to the logistic regression model (0.914, P=.382), exceeded that of MELD score (0.667, P<.001). The results were confirmed in the validation cohort. We have developed and validated a novel CART model superior to MELD for predicting three-month mortality of patients with ACHBLF. Thus, the CART model could facilitate medical decision-making and provide clinicians with a validated practical bedside tool for ACHBLF risk stratification. © 2016 John Wiley & Sons Ltd.

Confounder summary scores when comparing the effects of multiple drug exposures.

PubMed

Cadarette, Suzanne M; Gagne, Joshua J; Solomon, Daniel H; Katz, Jeffrey N; Stürmer, Til

2010-01-01

Little information is available comparing methods to adjust for confounding when considering multiple drug exposures. We compared three analytic strategies to control for confounding based on measured variables: conventional multivariable, exposure propensity score (EPS), and disease risk score (DRS). Each method was applied to a dataset (2000-2006) recently used to examine the comparative effectiveness of four drugs. The relative effectiveness of risedronate, nasal calcitonin, and raloxifene in preventing non-vertebral fracture, were each compared to alendronate. EPSs were derived both by using multinomial logistic regression (single model EPS) and by three separate logistic regression models (separate model EPS). DRSs were derived and event rates compared using Cox proportional hazard models. DRSs derived among the entire cohort (full cohort DRS) was compared to DRSs derived only among the referent alendronate (unexposed cohort DRS). Less than 8% deviation from the base estimate (conventional multivariable) was observed applying single model EPS, separate model EPS or full cohort DRS. Applying the unexposed cohort DRS when background risk for fracture differed between comparison drug exposure cohorts resulted in -7 to + 13% deviation from our base estimate. With sufficient numbers of exposed and outcomes, either conventional multivariable, EPS or full cohort DRS may be used to adjust for confounding to compare the effects of multiple drug exposures. However, our data also suggest that unexposed cohort DRS may be problematic when background risks differ between referent and exposed groups. Further empirical and simulation studies will help to clarify the generalizability of our findings.

An empirical model for estimating annual consumption by freshwater fish populations

USGS Publications Warehouse

Liao, H.; Pierce, C.L.; Larscheid, J.G.

2005-01-01

Population consumption is an important process linking predator populations to their prey resources. Simple tools are needed to enable fisheries managers to estimate population consumption. We assembled 74 individual estimates of annual consumption by freshwater fish populations and their mean annual population size, 41 of which also included estimates of mean annual biomass. The data set included 14 freshwater fish species from 10 different bodies of water. From this data set we developed two simple linear regression models predicting annual population consumption. Log-transformed population size explained 94% of the variation in log-transformed annual population consumption. Log-transformed biomass explained 98% of the variation in log-transformed annual population consumption. We quantified the accuracy of our regressions and three alternative consumption models as the mean percent difference from observed (bioenergetics-derived) estimates in a test data set. Predictions from our population-size regression matched observed consumption estimates poorly (mean percent difference = 222%). Predictions from our biomass regression matched observed consumption reasonably well (mean percent difference = 24%). The biomass regression was superior to an alternative model, similar in complexity, and comparable to two alternative models that were more complex and difficult to apply. Our biomass regression model, log10(consumption) = 0.5442 + 0.9962??log10(biomass), will be a useful tool for fishery managers, enabling them to make reasonably accurate annual population consumption predictions from mean annual biomass estimates. ?? Copyright by the American Fisheries Society 2005.

Mortality Predicted Accuracy for Hepatocellular Carcinoma Patients with Hepatic Resection Using Artificial Neural Network

PubMed Central

Chiu, Herng-Chia; Ho, Te-Wei; Lee, King-Teh; Chen, Hong-Yaw; Ho, Wen-Hsien

2013-01-01

The aim of this present study is firstly to compare significant predictors of mortality for hepatocellular carcinoma (HCC) patients undergoing resection between artificial neural network (ANN) and logistic regression (LR) models and secondly to evaluate the predictive accuracy of ANN and LR in different survival year estimation models. We constructed a prognostic model for 434 patients with 21 potential input variables by Cox regression model. Model performance was measured by numbers of significant predictors and predictive accuracy. The results indicated that ANN had double to triple numbers of significant predictors at 1-, 3-, and 5-year survival models as compared with LR models. Scores of accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) of 1-, 3-, and 5-year survival estimation models using ANN were superior to those of LR in all the training sets and most of the validation sets. The study demonstrated that ANN not only had a great number of predictors of mortality variables but also provided accurate prediction, as compared with conventional methods. It is suggested that physicians consider using data mining methods as supplemental tools for clinical decision-making and prognostic evaluation. PMID:23737707

PREDICTION OF MALIGNANT BREAST LESIONS FROM MRI FEATURES: A COMPARISON OF ARTIFICIAL NEURAL NETWORK AND LOGISTIC REGRESSION TECHNIQUES

PubMed Central

McLaren, Christine E.; Chen, Wen-Pin; Nie, Ke; Su, Min-Ying

2009-01-01

Rationale and Objectives Dynamic contrast enhanced MRI (DCE-MRI) is a clinical imaging modality for detection and diagnosis of breast lesions. Analytical methods were compared for diagnostic feature selection and performance of lesion classification to differentiate between malignant and benign lesions in patients. Materials and Methods The study included 43 malignant and 28 benign histologically-proven lesions. Eight morphological parameters, ten gray level co-occurrence matrices (GLCM) texture features, and fourteen Laws’ texture features were obtained using automated lesion segmentation and quantitative feature extraction. Artificial neural network (ANN) and logistic regression analysis were compared for selection of the best predictors of malignant lesions among the normalized features. Results Using ANN, the final four selected features were compactness, energy, homogeneity, and Law_LS, with area under the receiver operating characteristic curve (AUC) = 0.82, and accuracy = 0.76. The diagnostic performance of these 4-features computed on the basis of logistic regression yielded AUC = 0.80 (95% CI, 0.688 to 0.905), similar to that of ANN. The analysis also shows that the odds of a malignant lesion decreased by 48% (95% CI, 25% to 92%) for every increase of 1 SD in the Law_LS feature, adjusted for differences in compactness, energy, and homogeneity. Using logistic regression with z-score transformation, a model comprised of compactness, NRL entropy, and gray level sum average was selected, and it had the highest overall accuracy of 0.75 among all models, with AUC = 0.77 (95% CI, 0.660 to 0.880). When logistic modeling of transformations using the Box-Cox method was performed, the most parsimonious model with predictors, compactness and Law_LS, had an AUC of 0.79 (95% CI, 0.672 to 0.898). Conclusion The diagnostic performance of models selected by ANN and logistic regression was similar. The analytic methods were found to be roughly equivalent in terms of predictive ability when a small number of variables were chosen. The robust ANN methodology utilizes a sophisticated non-linear model, while logistic regression analysis provides insightful information to enhance interpretation of the model features. PMID:19409817

Improving near-infrared prediction model robustness with support vector machine regression: a pharmaceutical tablet assay example.

PubMed

Igne, Benoît; Drennen, James K; Anderson, Carl A

2014-01-01

Changes in raw materials and process wear and tear can have significant effects on the prediction error of near-infrared calibration models. When the variability that is present during routine manufacturing is not included in the calibration, test, and validation sets, the long-term performance and robustness of the model will be limited. Nonlinearity is a major source of interference. In near-infrared spectroscopy, nonlinearity can arise from light path-length differences that can come from differences in particle size or density. The usefulness of support vector machine (SVM) regression to handle nonlinearity and improve the robustness of calibration models in scenarios where the calibration set did not include all the variability present in test was evaluated. Compared to partial least squares (PLS) regression, SVM regression was less affected by physical (particle size) and chemical (moisture) differences. The linearity of the SVM predicted values was also improved. Nevertheless, although visualization and interpretation tools have been developed to enhance the usability of SVM-based methods, work is yet to be done to provide chemometricians in the pharmaceutical industry with a regression method that can supplement PLS-based methods.

Noninvasive and fast measurement of blood glucose in vivo by near infrared (NIR) spectroscopy

NASA Astrophysics Data System (ADS)

Jintao, Xue; Liming, Ye; Yufei, Liu; Chunyan, Li; Han, Chen

2017-05-01

This research was to develop a method for noninvasive and fast blood glucose assay in vivo. Near-infrared (NIR) spectroscopy, a more promising technique compared to other methods, was investigated in rats with diabetes and normal rats. Calibration models are generated by two different multivariate strategies: partial least squares (PLS) as linear regression method and artificial neural networks (ANN) as non-linear regression method. The PLS model was optimized individually by considering spectral range, spectral pretreatment methods and number of model factors, while the ANN model was studied individually by selecting spectral pretreatment methods, parameters of network topology, number of hidden neurons, and times of epoch. The results of the validation showed the two models were robust, accurate and repeatable. Compared to the ANN model, the performance of the PLS model was much better, with lower root mean square error of validation (RMSEP) of 0.419 and higher correlation coefficients (R) of 96.22%.

Sperm Retrieval in Patients with Klinefelter Syndrome: A Skewed Regression Model Analysis.

PubMed

Chehrazi, Mohammad; Rahimiforoushani, Abbas; Sabbaghian, Marjan; Nourijelyani, Keramat; Sadighi Gilani, Mohammad Ali; Hoseini, Mostafa; Vesali, Samira; Yaseri, Mehdi; Alizadeh, Ahad; Mohammad, Kazem; Samani, Reza Omani

2017-01-01

The most common chromosomal abnormality due to non-obstructive azoospermia (NOA) is Klinefelter syndrome (KS) which occurs in 1-1.72 out of 500-1000 male infants. The probability of retrieving sperm as the outcome could be asymmetrically different between patients with and without KS, therefore logistic regression analysis is not a well-qualified test for this type of data. This study has been designed to evaluate skewed regression model analysis for data collected from microsurgical testicular sperm extraction (micro-TESE) among azoospermic patients with and without non-mosaic KS syndrome. This cohort study compared the micro-TESE outcome between 134 men with classic KS and 537 men with NOA and normal karyotype who were referred to Royan Institute between 2009 and 2011. In addition to our main outcome, which was sperm retrieval, we also used logistic and skewed regression analyses to compare the following demographic and hormonal factors: age, level of follicle stimulating hormone (FSH), luteinizing hormone (LH), and testosterone between the two groups. A comparison of the micro-TESE between the KS and control groups showed a success rate of 28.4% (38/134) for the KS group and 22.2% (119/537) for the control group. In the KS group, a significantly difference (P<0.001) existed between testosterone levels for the successful sperm retrieval group (3.4 ± 0.48 mg/mL) compared to the unsuccessful sperm retrieval group (2.33 ± 0.23 mg/mL). The index for quasi Akaike information criterion (QAIC) had a goodness of fit of 74 for the skewed model which was lower than logistic regression (QAIC=85). According to the results, skewed regression is more efficient in estimating sperm retrieval success when the data from patients with KS are analyzed. This finding should be investigated by conducting additional studies with different data structures.

Comparing spatially varying coefficient models: a case study examining violent crime rates and their relationships to alcohol outlets and illegal drug arrests

NASA Astrophysics Data System (ADS)

Wheeler, David C.; Waller, Lance A.

2009-03-01

In this paper, we compare and contrast a Bayesian spatially varying coefficient process (SVCP) model with a geographically weighted regression (GWR) model for the estimation of the potentially spatially varying regression effects of alcohol outlets and illegal drug activity on violent crime in Houston, Texas. In addition, we focus on the inherent coefficient shrinkage properties of the Bayesian SVCP model as a way to address increased coefficient variance that follows from collinearity in GWR models. We outline the advantages of the Bayesian model in terms of reducing inflated coefficient variance, enhanced model flexibility, and more formal measuring of model uncertainty for prediction. We find spatially varying effects for alcohol outlets and drug violations, but the amount of variation depends on the type of model used. For the Bayesian model, this variation is controllable through the amount of prior influence placed on the variance of the coefficients. For example, the spatial pattern of coefficients is similar for the GWR and Bayesian models when a relatively large prior variance is used in the Bayesian model.

A simple linear regression method for quantitative trait loci linkage analysis with censored observations.

PubMed

Anderson, Carl A; McRae, Allan F; Visscher, Peter M

2006-07-01

Standard quantitative trait loci (QTL) mapping techniques commonly assume that the trait is both fully observed and normally distributed. When considering survival or age-at-onset traits these assumptions are often incorrect. Methods have been developed to map QTL for survival traits; however, they are both computationally intensive and not available in standard genome analysis software packages. We propose a grouped linear regression method for the analysis of continuous survival data. Using simulation we compare this method to both the Cox and Weibull proportional hazards models and a standard linear regression method that ignores censoring. The grouped linear regression method is of equivalent power to both the Cox and Weibull proportional hazards methods and is significantly better than the standard linear regression method when censored observations are present. The method is also robust to the proportion of censored individuals and the underlying distribution of the trait. On the basis of linear regression methodology, the grouped linear regression model is computationally simple and fast and can be implemented readily in freely available statistical software.

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

An Application to the Prediction of LOD Change Based on General Regression Neural Network

NASA Astrophysics Data System (ADS)

Zhang, X. H.; Wang, Q. J.; Zhu, J. J.; Zhang, H.

2011-07-01

Traditional prediction of the LOD (length of day) change was based on linear models, such as the least square model and the autoregressive technique, etc. Due to the complex non-linear features of the LOD variation, the performances of the linear model predictors are not fully satisfactory. This paper applies a non-linear neural network - general regression neural network (GRNN) model to forecast the LOD change, and the results are analyzed and compared with those obtained with the back propagation neural network and other models. The comparison shows that the performance of the GRNN model in the prediction of the LOD change is efficient and feasible.

Modelling nitrate pollution pressure using a multivariate statistical approach: the case of Kinshasa groundwater body, Democratic Republic of Congo

NASA Astrophysics Data System (ADS)

Mfumu Kihumba, Antoine; Ndembo Longo, Jean; Vanclooster, Marnik

2016-03-01

A multivariate statistical modelling approach was applied to explain the anthropogenic pressure of nitrate pollution on the Kinshasa groundwater body (Democratic Republic of Congo). Multiple regression and regression tree models were compared and used to identify major environmental factors that control the groundwater nitrate concentration in this region. The analyses were made in terms of physical attributes related to the topography, land use, geology and hydrogeology in the capture zone of different groundwater sampling stations. For the nitrate data, groundwater datasets from two different surveys were used. The statistical models identified the topography, the residential area, the service land (cemetery), and the surface-water land-use classes as major factors explaining nitrate occurrence in the groundwater. Also, groundwater nitrate pollution depends not on one single factor but on the combined influence of factors representing nitrogen loading sources and aquifer susceptibility characteristics. The groundwater nitrate pressure was better predicted with the regression tree model than with the multiple regression model. Furthermore, the results elucidated the sensitivity of the model performance towards the method of delineation of the capture zones. For pollution modelling at the monitoring points, therefore, it is better to identify capture-zone shapes based on a conceptual hydrogeological model rather than to adopt arbitrary circular capture zones.

The prediction of intelligence in preschool children using alternative models to regression.

PubMed

Finch, W Holmes; Chang, Mei; Davis, Andrew S; Holden, Jocelyn E; Rothlisberg, Barbara A; McIntosh, David E

2011-12-01

Statistical prediction of an outcome variable using multiple independent variables is a common practice in the social and behavioral sciences. For example, neuropsychologists are sometimes called upon to provide predictions of preinjury cognitive functioning for individuals who have suffered a traumatic brain injury. Typically, these predictions are made using standard multiple linear regression models with several demographic variables (e.g., gender, ethnicity, education level) as predictors. Prior research has shown conflicting evidence regarding the ability of such models to provide accurate predictions of outcome variables such as full-scale intelligence (FSIQ) test scores. The present study had two goals: (1) to demonstrate the utility of a set of alternative prediction methods that have been applied extensively in the natural sciences and business but have not been frequently explored in the social sciences and (2) to develop models that can be used to predict premorbid cognitive functioning in preschool children. Predictions of Stanford-Binet 5 FSIQ scores for preschool-aged children is used to compare the performance of a multiple regression model with several of these alternative methods. Results demonstrate that classification and regression trees provided more accurate predictions of FSIQ scores than does the more traditional regression approach. Implications of these results are discussed.

Regional regression of flood characteristics employing historical information

USGS Publications Warehouse

Tasker, Gary D.; Stedinger, J.R.

1987-01-01

Streamflow gauging networks provide hydrologic information for use in estimating the parameters of regional regression models. The regional regression models can be used to estimate flood statistics, such as the 100 yr peak, at ungauged sites as functions of drainage basin characteristics. A recent innovation in regional regression is the use of a generalized least squares (GLS) estimator that accounts for unequal station record lengths and sample cross correlation among the flows. However, this technique does not account for historical flood information. A method is proposed here to adjust this generalized least squares estimator to account for possible information about historical floods available at some stations in a region. The historical information is assumed to be in the form of observations of all peaks above a threshold during a long period outside the systematic record period. A Monte Carlo simulation experiment was performed to compare the GLS estimator adjusted for historical floods with the unadjusted GLS estimator and the ordinary least squares estimator. Results indicate that using the GLS estimator adjusted for historical information significantly improves the regression model. ?? 1987.

Linear regression models for solvent accessibility prediction in proteins.

PubMed

Wagner, Michael; Adamczak, Rafał; Porollo, Aleksey; Meller, Jarosław

2005-04-01

The relative solvent accessibility (RSA) of an amino acid residue in a protein structure is a real number that represents the solvent exposed surface area of this residue in relative terms. The problem of predicting the RSA from the primary amino acid sequence can therefore be cast as a regression problem. Nevertheless, RSA prediction has so far typically been cast as a classification problem. Consequently, various machine learning techniques have been used within the classification framework to predict whether a given amino acid exceeds some (arbitrary) RSA threshold and would thus be predicted to be "exposed," as opposed to "buried." We have recently developed novel methods for RSA prediction using nonlinear regression techniques which provide accurate estimates of the real-valued RSA and outperform classification-based approaches with respect to commonly used two-class projections. However, while their performance seems to provide a significant improvement over previously published approaches, these Neural Network (NN) based methods are computationally expensive to train and involve several thousand parameters. In this work, we develop alternative regression models for RSA prediction which are computationally much less expensive, involve orders-of-magnitude fewer parameters, and are still competitive in terms of prediction quality. In particular, we investigate several regression models for RSA prediction using linear L1-support vector regression (SVR) approaches as well as standard linear least squares (LS) regression. Using rigorously derived validation sets of protein structures and extensive cross-validation analysis, we compare the performance of the SVR with that of LS regression and NN-based methods. In particular, we show that the flexibility of the SVR (as encoded by metaparameters such as the error insensitivity and the error penalization terms) can be very beneficial to optimize the prediction accuracy for buried residues. We conclude that the simple and computationally much more efficient linear SVR performs comparably to nonlinear models and thus can be used in order to facilitate further attempts to design more accurate RSA prediction methods, with applications to fold recognition and de novo protein structure prediction methods.

Content Coding of Psychotherapy Transcripts Using Labeled Topic Models.

PubMed

Gaut, Garren; Steyvers, Mark; Imel, Zac E; Atkins, David C; Smyth, Padhraic

2017-03-01

Psychotherapy represents a broad class of medical interventions received by millions of patients each year. Unlike most medical treatments, its primary mechanisms are linguistic; i.e., the treatment relies directly on a conversation between a patient and provider. However, the evaluation of patient-provider conversation suffers from critical shortcomings, including intensive labor requirements, coder error, nonstandardized coding systems, and inability to scale up to larger data sets. To overcome these shortcomings, psychotherapy analysis needs a reliable and scalable method for summarizing the content of treatment encounters. We used a publicly available psychotherapy corpus from Alexander Street press comprising a large collection of transcripts of patient-provider conversations to compare coding performance for two machine learning methods. We used the labeled latent Dirichlet allocation (L-LDA) model to learn associations between text and codes, to predict codes in psychotherapy sessions, and to localize specific passages of within-session text representative of a session code. We compared the L-LDA model to a baseline lasso regression model using predictive accuracy and model generalizability (measured by calculating the area under the curve (AUC) from the receiver operating characteristic curve). The L-LDA model outperforms the lasso logistic regression model at predicting session-level codes with average AUC scores of 0.79, and 0.70, respectively. For fine-grained level coding, L-LDA and logistic regression are able to identify specific talk-turns representative of symptom codes. However, model performance for talk-turn identification is not yet as reliable as human coders. We conclude that the L-LDA model has the potential to be an objective, scalable method for accurate automated coding of psychotherapy sessions that perform better than comparable discriminative methods at session-level coding and can also predict fine-grained codes.

Content Coding of Psychotherapy Transcripts Using Labeled Topic Models

PubMed Central

Gaut, Garren; Steyvers, Mark; Imel, Zac E; Atkins, David C; Smyth, Padhraic

2016-01-01

Psychotherapy represents a broad class of medical interventions received by millions of patients each year. Unlike most medical treatments, its primary mechanisms are linguistic; i.e., the treatment relies directly on a conversation between a patient and provider. However, the evaluation of patient-provider conversation suffers from critical shortcomings, including intensive labor requirements, coder error, non-standardized coding systems, and inability to scale up to larger data sets. To overcome these shortcomings, psychotherapy analysis needs a reliable and scalable method for summarizing the content of treatment encounters. We used a publicly-available psychotherapy corpus from Alexander Street press comprising a large collection of transcripts of patient-provider conversations to compare coding performance for two machine learning methods. We used the Labeled Latent Dirichlet Allocation (L-LDA) model to learn associations between text and codes, to predict codes in psychotherapy sessions, and to localize specific passages of within-session text representative of a session code. We compared the L-LDA model to a baseline lasso regression model using predictive accuracy and model generalizability (measured by calculating the area under the curve (AUC) from the receiver operating characteristic (ROC) curve). The L-LDA model outperforms the lasso logistic regression model at predicting session-level codes with average AUC scores of .79, and .70, respectively. For fine-grained level coding, L-LDA and logistic regression are able to identify specific talk-turns representative of symptom codes. However, model performance for talk-turn identification is not yet as reliable as human coders. We conclude that the L-LDA model has the potential to be an objective, scaleable method for accurate automated coding of psychotherapy sessions that performs better than comparable discriminative methods at session-level coding and can also predict fine-grained codes. PMID:26625437

Robust and efficient estimation with weighted composite quantile regression

NASA Astrophysics Data System (ADS)

Jiang, Xuejun; Li, Jingzhi; Xia, Tian; Yan, Wanfeng

2016-09-01

In this paper we introduce a weighted composite quantile regression (CQR) estimation approach and study its application in nonlinear models such as exponential models and ARCH-type models. The weighted CQR is augmented by using a data-driven weighting scheme. With the error distribution unspecified, the proposed estimators share robustness from quantile regression and achieve nearly the same efficiency as the oracle maximum likelihood estimator (MLE) for a variety of error distributions including the normal, mixed-normal, Student's t, Cauchy distributions, etc. We also suggest an algorithm for the fast implementation of the proposed methodology. Simulations are carried out to compare the performance of different estimators, and the proposed approach is used to analyze the daily S&P 500 Composite index, which verifies the effectiveness and efficiency of our theoretical results.

A comparison of Cox and logistic regression for use in genome-wide association studies of cohort and case-cohort design.

PubMed

Staley, James R; Jones, Edmund; Kaptoge, Stephen; Butterworth, Adam S; Sweeting, Michael J; Wood, Angela M; Howson, Joanna M M

2017-06-01

Logistic regression is often used instead of Cox regression to analyse genome-wide association studies (GWAS) of single-nucleotide polymorphisms (SNPs) and disease outcomes with cohort and case-cohort designs, as it is less computationally expensive. Although Cox and logistic regression models have been compared previously in cohort studies, this work does not completely cover the GWAS setting nor extend to the case-cohort study design. Here, we evaluated Cox and logistic regression applied to cohort and case-cohort genetic association studies using simulated data and genetic data from the EPIC-CVD study. In the cohort setting, there was a modest improvement in power to detect SNP-disease associations using Cox regression compared with logistic regression, which increased as the disease incidence increased. In contrast, logistic regression had more power than (Prentice weighted) Cox regression in the case-cohort setting. Logistic regression yielded inflated effect estimates (assuming the hazard ratio is the underlying measure of association) for both study designs, especially for SNPs with greater effect on disease. Given logistic regression is substantially more computationally efficient than Cox regression in both settings, we propose a two-step approach to GWAS in cohort and case-cohort studies. First to analyse all SNPs with logistic regression to identify associated variants below a pre-defined P-value threshold, and second to fit Cox regression (appropriately weighted in case-cohort studies) to those identified SNPs to ensure accurate estimation of association with disease.

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.

Application of Boosting Regression Trees to Preliminary Cost Estimation in Building Construction Projects

PubMed Central

2015-01-01

Among the recent data mining techniques available, the boosting approach has attracted a great deal of attention because of its effective learning algorithm and strong boundaries in terms of its generalization performance. However, the boosting approach has yet to be used in regression problems within the construction domain, including cost estimations, but has been actively utilized in other domains. Therefore, a boosting regression tree (BRT) is applied to cost estimations at the early stage of a construction project to examine the applicability of the boosting approach to a regression problem within the construction domain. To evaluate the performance of the BRT model, its performance was compared with that of a neural network (NN) model, which has been proven to have a high performance in cost estimation domains. The BRT model has shown results similar to those of NN model using 234 actual cost datasets of a building construction project. In addition, the BRT model can provide additional information such as the importance plot and structure model, which can support estimators in comprehending the decision making process. Consequently, the boosting approach has potential applicability in preliminary cost estimations in a building construction project. PMID:26339227

Application of Boosting Regression Trees to Preliminary Cost Estimation in Building Construction Projects.

PubMed

Shin, Yoonseok

2015-01-01

Among the recent data mining techniques available, the boosting approach has attracted a great deal of attention because of its effective learning algorithm and strong boundaries in terms of its generalization performance. However, the boosting approach has yet to be used in regression problems within the construction domain, including cost estimations, but has been actively utilized in other domains. Therefore, a boosting regression tree (BRT) is applied to cost estimations at the early stage of a construction project to examine the applicability of the boosting approach to a regression problem within the construction domain. To evaluate the performance of the BRT model, its performance was compared with that of a neural network (NN) model, which has been proven to have a high performance in cost estimation domains. The BRT model has shown results similar to those of NN model using 234 actual cost datasets of a building construction project. In addition, the BRT model can provide additional information such as the importance plot and structure model, which can support estimators in comprehending the decision making process. Consequently, the boosting approach has potential applicability in preliminary cost estimations in a building construction project.

USING HYDROGRAPHIC DATA AND THE EPA VIRTUAL BEACH MODEL TO TEST PREDICTIONS OF BEACH BACTERIA CONCENTRATIONS

EPA Science Inventory

A modeling study of 2006 Huntington Beach (Lake Erie) beach bacteria concentrations indicates multi-variable linear regression (MLR) can effectively estimate bacteria concentrations compared to the persistence model. Our use of the Virtual Beach (VB) model affirms that fact. VB i...

Development of hybrid genetic-algorithm-based neural networks using regression trees for modeling air quality inside a public transportation bus.

PubMed

Kadiyala, Akhil; Kaur, Devinder; Kumar, Ashok

2013-02-01

The present study developed a novel approach to modeling indoor air quality (IAQ) of a public transportation bus by the development of hybrid genetic-algorithm-based neural networks (also known as evolutionary neural networks) with input variables optimized from using the regression trees, referred as the GART approach. This study validated the applicability of the GART modeling approach in solving complex nonlinear systems by accurately predicting the monitored contaminants of carbon dioxide (CO2), carbon monoxide (CO), nitric oxide (NO), sulfur dioxide (SO2), 0.3-0.4 microm sized particle numbers, 0.4-0.5 microm sized particle numbers, particulate matter (PM) concentrations less than 1.0 microm (PM10), and PM concentrations less than 2.5 microm (PM2.5) inside a public transportation bus operating on 20% grade biodiesel in Toledo, OH. First, the important variables affecting each monitored in-bus contaminant were determined using regression trees. Second, the analysis of variance was used as a complimentary sensitivity analysis to the regression tree results to determine a subset of statistically significant variables affecting each monitored in-bus contaminant. Finally, the identified subsets of statistically significant variables were used as inputs to develop three artificial neural network (ANN) models. The models developed were regression tree-based back-propagation network (BPN-RT), regression tree-based radial basis function network (RBFN-RT), and GART models. Performance measures were used to validate the predictive capacity of the developed IAQ models. The results from this approach were compared with the results obtained from using a theoretical approach and a generalized practicable approach to modeling IAQ that included the consideration of additional independent variables when developing the aforementioned ANN models. The hybrid GART models were able to capture majority of the variance in the monitored in-bus contaminants. The genetic-algorithm-based neural network IAQ models outperformed the traditional ANN methods of the back-propagation and the radial basis function networks. The novelty of this research is the development of a novel approach to modeling vehicular indoor air quality by integration of the advanced methods of genetic algorithms, regression trees, and the analysis of variance for the monitored in-vehicle gaseous and particulate matter contaminants, and comparing the results obtained from using the developed approach with conventional artificial intelligence techniques of back propagation networks and radial basis function networks. This study validated the newly developed approach using holdout and threefold cross-validation methods. These results are of great interest to scientists, researchers, and the public in understanding the various aspects of modeling an indoor microenvironment. This methodology can easily be extended to other fields of study also.

Numerical investigation on the regression rate of hybrid rocket motor with star swirl fuel grain

NASA Astrophysics Data System (ADS)

Zhang, Shuai; Hu, Fan; Zhang, Weihua

2016-10-01

Although hybrid rocket motor is prospected to have distinct advantages over liquid and solid rocket motor, low regression rate and insufficient efficiency are two major disadvantages which have prevented it from being commercially viable. In recent years, complex fuel grain configurations are attractive in overcoming the disadvantages with the help of Rapid Prototyping technology. In this work, an attempt has been made to numerically investigate the flow field characteristics and local regression rate distribution inside the hybrid rocket motor with complex star swirl grain. A propellant combination with GOX and HTPB has been chosen. The numerical model is established based on the three dimensional Navier-Stokes equations with turbulence, combustion, and coupled gas/solid phase formulations. The calculated fuel regression rate is compared with the experimental data to validate the accuracy of numerical model. The results indicate that, comparing the star swirl grain with the tube grain under the conditions of the same port area and the same grain length, the burning surface area rises about 200%, the spatially averaged regression rate rises as high as about 60%, and the oxidizer can combust sufficiently due to the big vortex around the axis in the aft-mixing chamber. The combustion efficiency of star swirl grain is better and more stable than that of tube grain.

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.

Calibration transfer of a Raman spectroscopic quantification method for the assessment of liquid detergent compositions from at-line laboratory to in-line industrial scale.

PubMed

Brouckaert, D; Uyttersprot, J-S; Broeckx, W; De Beer, T

2018-03-01

Calibration transfer or standardisation aims at creating a uniform spectral response on different spectroscopic instruments or under varying conditions, without requiring a full recalibration for each situation. In the current study, this strategy is applied to construct at-line multivariate calibration models and consequently employ them in-line in a continuous industrial production line, using the same spectrometer. Firstly, quantitative multivariate models are constructed at-line at laboratory scale for predicting the concentration of two main ingredients in hard surface cleaners. By regressing the Raman spectra of a set of small-scale calibration samples against their reference concentration values, partial least squares (PLS) models are developed to quantify the surfactant levels in the liquid detergent compositions under investigation. After evaluating the models performance with a set of independent validation samples, a univariate slope/bias correction is applied in view of transporting these at-line calibration models to an in-line manufacturing set-up. This standardisation technique allows a fast and easy transfer of the PLS regression models, by simply correcting the model predictions on the in-line set-up, without adjusting anything to the original multivariate calibration models. An extensive statistical analysis is performed in order to assess the predictive quality of the transferred regression models. Before and after transfer, the R 2 and RMSEP of both models is compared for evaluating if their magnitude is similar. T-tests are then performed to investigate whether the slope and intercept of the transferred regression line are not statistically different from 1 and 0, respectively. Furthermore, it is inspected whether no significant bias can be noted. F-tests are executed as well, for assessing the linearity of the transfer regression line and for investigating the statistical coincidence of the transfer and validation regression line. Finally, a paired t-test is performed to compare the original at-line model to the slope/bias corrected in-line model, using interval hypotheses. It is shown that the calibration models of Surfactant 1 and Surfactant 2 yield satisfactory in-line predictions after slope/bias correction. While Surfactant 1 passes seven out of eight statistical tests, the recommended validation parameters are 100% successful for Surfactant 2. It is hence concluded that the proposed strategy for transferring at-line calibration models to an in-line industrial environment via a univariate slope/bias correction of the predicted values offers a successful standardisation approach. Copyright © 2017 Elsevier B.V. All rights reserved.

Modeling and forecasting US presidential election using learning algorithms

NASA Astrophysics Data System (ADS)

Zolghadr, Mohammad; Niaki, Seyed Armin Akhavan; Niaki, S. T. A.

2017-09-01

The primary objective of this research is to obtain an accurate forecasting model for the US presidential election. To identify a reliable model, artificial neural networks (ANN) and support vector regression (SVR) models are compared based on some specified performance measures. Moreover, six independent variables such as GDP, unemployment rate, the president's approval rate, and others are considered in a stepwise regression to identify significant variables. The president's approval rate is identified as the most significant variable, based on which eight other variables are identified and considered in the model development. Preprocessing methods are applied to prepare the data for the learning algorithms. The proposed procedure significantly increases the accuracy of the model by 50%. The learning algorithms (ANN and SVR) proved to be superior to linear regression based on each method's calculated performance measures. The SVR model is identified as the most accurate model among the other models as this model successfully predicted the outcome of the election in the last three elections (2004, 2008, and 2012). The proposed approach significantly increases the accuracy of the forecast.

Global-scale high-resolution ( 1 km) modelling of mean, maximum and minimum annual streamflow

NASA Astrophysics Data System (ADS)

Barbarossa, Valerio; Huijbregts, Mark; Hendriks, Jan; Beusen, Arthur; Clavreul, Julie; King, Henry; Schipper, Aafke

2017-04-01

Quantifying mean, maximum and minimum annual flow (AF) of rivers at ungauged sites is essential for a number of applications, including assessments of global water supply, ecosystem integrity and water footprints. AF metrics can be quantified with spatially explicit process-based models, which might be overly time-consuming and data-intensive for this purpose, or with empirical regression models that predict AF metrics based on climate and catchment characteristics. Yet, so far, regression models have mostly been developed at a regional scale and the extent to which they can be extrapolated to other regions is not known. We developed global-scale regression models that quantify mean, maximum and minimum AF as function of catchment area and catchment-averaged slope, elevation, and mean, maximum and minimum annual precipitation and air temperature. We then used these models to obtain global 30 arc-seconds (˜ 1 km) maps of mean, maximum and minimum AF for each year from 1960 through 2015, based on a newly developed hydrologically conditioned digital elevation model. We calibrated our regression models based on observations of discharge and catchment characteristics from about 4,000 catchments worldwide, ranging from 100 to 106 km2 in size, and validated them against independent measurements as well as the output of a number of process-based global hydrological models (GHMs). The variance explained by our regression models ranged up to 90% and the performance of the models compared well with the performance of existing GHMs. Yet, our AF maps provide a level of spatial detail that cannot yet be achieved by current GHMs.

Application of Hierarchical Linear Models/Linear Mixed-Effects Models in School Effectiveness Research

ERIC Educational Resources Information Center

Ker, H. W.

2014-01-01

Multilevel data are very common in educational research. Hierarchical linear models/linear mixed-effects models (HLMs/LMEs) are often utilized to analyze multilevel data nowadays. This paper discusses the problems of utilizing ordinary regressions for modeling multilevel educational data, compare the data analytic results from three regression…

System dynamic modeling: an alternative method for budgeting.

PubMed

Srijariya, Witsanuchai; Riewpaiboon, Arthorn; Chaikledkaew, Usa

2008-03-01

To construct, validate, and simulate a system dynamic financial model and compare it against the conventional method. The study was a cross-sectional analysis of secondary data retrieved from the National Health Security Office (NHSO) in the fiscal year 2004. The sample consisted of all emergency patients who received emergency services outside their registered hospital-catchments area. The dependent variable used was the amount of reimbursed money. Two types of model were constructed, namely, the system dynamic model using the STELLA software and the multiple linear regression model. The outputs of both methods were compared. The study covered 284,716 patients from various levels of providers. The system dynamic model had the capability of producing various types of outputs, for example, financial and graphical analyses. For the regression analysis, statistically significant predictors were composed of service types (outpatient or inpatient), operating procedures, length of stay, illness types (accident or not), hospital characteristics, age, and hospital location (adjusted R(2) = 0.74). The total budget arrived at from using the system dynamic model and regression model was US$12,159,614.38 and US$7,301,217.18, respectively, whereas the actual NHSO reimbursement cost was US$12,840,805.69. The study illustrated that the system dynamic model is a useful financial management tool, although it is not easy to construct. The model is not only more accurate in prediction but is also more capable of analyzing large and complex real-world situations than the conventional method.

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…

Estimating irrigation water use in the humid eastern United States

USGS Publications Warehouse

Levin, Sara B.; Zarriello, Phillip J.

2013-01-01

Accurate accounting of irrigation water use is an important part of the U.S. Geological Survey National Water-Use Information Program and the WaterSMART initiative to help maintain sustainable water resources in the Nation. Irrigation water use in the humid eastern United States is not well characterized because of inadequate reporting and wide variability associated with climate, soils, crops, and farming practices. To better understand irrigation water use in the eastern United States, two types of predictive models were developed and compared by using metered irrigation water-use data for corn, cotton, peanut, and soybean crops in Georgia and turf farms in Rhode Island. Reliable metered irrigation data were limited to these areas. The first predictive model that was developed uses logistic regression to predict the occurrence of irrigation on the basis of antecedent climate conditions. Logistic regression equations were developed for corn, cotton, peanut, and soybean crops by using weekly irrigation water-use data from 36 metered sites in Georgia in 2009 and 2010 and turf farms in Rhode Island from 2000 to 2004. For the weeks when irrigation was predicted to take place, the irrigation water-use volume was estimated by multiplying the average metered irrigation application rate by the irrigated acreage for a given crop. The second predictive model that was developed is a crop-water-demand model that uses a daily soil water balance to estimate the water needs of a crop on a given day based on climate, soil, and plant properties. Crop-water-demand models were developed independently of reported irrigation water-use practices and relied on knowledge of plant properties that are available in the literature. Both modeling approaches require accurate accounting of irrigated area and crop type to estimate total irrigation water use. Water-use estimates from both modeling methods were compared to the metered irrigation data from Rhode Island and Georgia that were used to develop the models as well as two independent validation datasets from Georgia and Virginia that were not used in model development. Irrigation water-use estimates from the logistic regression method more closely matched mean reported irrigation rates than estimates from the crop-water-demand model when compared to the irrigation data used to develop the equations. The root mean squared errors (RMSEs) for the logistic regression estimates of mean annual irrigation ranged from 0.3 to 2.0 inches (in.) for the five crop types; RMSEs for the crop-water-demand models ranged from 1.4 to 3.9 in. However, when the models were applied and compared to the independent validation datasets from southwest Georgia from 2010, and from Virginia from 1999 to 2007, the crop-water-demand model estimates were as good as or better at predicting the mean irrigation volume than the logistic regression models for most crop types. RMSEs for logistic regression estimates of mean annual irrigation ranged from 1.0 to 7.0 in. for validation data from Georgia and from 1.8 to 4.9 in. for validation data from Virginia; RMSEs for crop-water-demand model estimates ranged from 2.1 to 5.8 in. for Georgia data and from 2.0 to 3.9 in. for Virginia data. In general, regression-based models performed better in areas that had quality daily or weekly irrigation data from which the regression equations were developed; however, the regression models were less reliable than the crop-water-demand models when applied outside the area for which they were developed. In most eastern coastal states that do not have quality irrigation data, the crop-water-demand model can be used more reliably. The development of predictive models of irrigation water use in this study was hindered by a lack of quality irrigation data. Many mid-Atlantic and New England states do not require irrigation water use to be reported. A survey of irrigation data from 14 eastern coastal states from Maine to Georgia indicated that, with the exception of the data in Georgia, irrigation data in the states that do require reporting commonly did not contain requisite ancillary information such as irrigated area or crop type, lacked precision, or were at an aggregated temporal scale making them unsuitable for use in the development of predictive models. Confidence in the reliability of either modeling method is affected by uncertainty in the reported data from which the models were developed or validated. Only through additional collection of quality data and further study can the accuracy and uncertainty of irrigation water-use estimates be improved in the humid eastern United States.

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

Simulation of parametric model towards the fixed covariate of right censored lung cancer data

NASA Astrophysics Data System (ADS)

Afiqah Muhamad Jamil, Siti; Asrul Affendi Abdullah, M.; Kek, Sie Long; Ridwan Olaniran, Oyebayo; Enera Amran, Syahila

2017-09-01

In this study, simulation procedure was applied to measure the fixed covariate of right censored data by using parametric survival model. The scale and shape parameter were modified to differentiate the analysis of parametric regression survival model. Statistically, the biases, mean biases and the coverage probability were used in this analysis. Consequently, different sample sizes were employed to distinguish the impact of parametric regression model towards right censored data with 50, 100, 150 and 200 number of sample. R-statistical software was utilised to develop the coding simulation with right censored data. Besides, the final model of right censored simulation was compared with the right censored lung cancer data in Malaysia. It was found that different values of shape and scale parameter with different sample size, help to improve the simulation strategy for right censored data and Weibull regression survival model is suitable fit towards the simulation of survival of lung cancer patients data in Malaysia.

On the use and misuse of scalar scores of confounders in design and analysis of observational studies.

PubMed

Pfeiffer, R M; Riedl, R

2015-08-15

We assess the asymptotic bias of estimates of exposure effects conditional on covariates when summary scores of confounders, instead of the confounders themselves, are used to analyze observational data. First, we study regression models for cohort data that are adjusted for summary scores. Second, we derive the asymptotic bias for case-control studies when cases and controls are matched on a summary score, and then analyzed either using conditional logistic regression or by unconditional logistic regression adjusted for the summary score. Two scores, the propensity score (PS) and the disease risk score (DRS) are studied in detail. For cohort analysis, when regression models are adjusted for the PS, the estimated conditional treatment effect is unbiased only for linear models, or at the null for non-linear models. Adjustment of cohort data for DRS yields unbiased estimates only for linear regression; all other estimates of exposure effects are biased. Matching cases and controls on DRS and analyzing them using conditional logistic regression yields unbiased estimates of exposure effect, whereas adjusting for the DRS in unconditional logistic regression yields biased estimates, even under the null hypothesis of no association. Matching cases and controls on the PS yield unbiased estimates only under the null for both conditional and unconditional logistic regression, adjusted for the PS. We study the bias for various confounding scenarios and compare our asymptotic results with those from simulations with limited sample sizes. To create realistic correlations among multiple confounders, we also based simulations on a real dataset. Copyright © 2015 John Wiley & Sons, Ltd.

An Analysis of San Diego's Housing Market Using a Geographically Weighted Regression Approach

NASA Astrophysics Data System (ADS)

Grant, Christina P.

San Diego County real estate transaction data was evaluated with a set of linear models calibrated by ordinary least squares and geographically weighted regression (GWR). The goal of the analysis was to determine whether the spatial effects assumed to be in the data are best studied globally with no spatial terms, globally with a fixed effects submarket variable, or locally with GWR. 18,050 single-family residential sales which closed in the six months between April 2014 and September 2014 were used in the analysis. Diagnostic statistics including AICc, R2, Global Moran's I, and visual inspection of diagnostic plots and maps indicate superior model performance by GWR as compared to both global regressions.

Application of General Regression Neural Network to the Prediction of LOD Change

NASA Astrophysics Data System (ADS)

Zhang, Xiao-Hong; Wang, Qi-Jie; Zhu, Jian-Jun; Zhang, Hao

2012-01-01

Traditional methods for predicting the change in length of day (LOD change) are mainly based on some linear models, such as the least square model and autoregression model, etc. However, the LOD change comprises complicated non-linear factors and the prediction effect of the linear models is always not so ideal. Thus, a kind of non-linear neural network — general regression neural network (GRNN) model is tried to make the prediction of the LOD change and the result is compared with the predicted results obtained by taking advantage of the BP (back propagation) neural network model and other models. The comparison result shows that the application of the GRNN to the prediction of the LOD change is highly effective and feasible.

Deciphering factors controlling groundwater arsenic spatial variability in Bangladesh

NASA Astrophysics Data System (ADS)

Tan, Z.; Yang, Q.; Zheng, C.; Zheng, Y.

2017-12-01

Elevated concentrations of geogenic arsenic in groundwater have been found in many countries to exceed 10 μg/L, the WHO's guideline value for drinking water. A common yet unexplained characteristic of groundwater arsenic spatial distribution is the extensive variability at various spatial scales. This study investigates factors influencing the spatial variability of groundwater arsenic in Bangladesh to improve the accuracy of models predicting arsenic exceedance rate spatially. A novel boosted regression tree method is used to establish a weak-learning ensemble model, which is compared to a linear model using a conventional stepwise logistic regression method. The boosted regression tree models offer the advantage of parametric interaction when big datasets are analyzed in comparison to the logistic regression. The point data set (n=3,538) of groundwater hydrochemistry with 19 parameters was obtained by the British Geological Survey in 2001. The spatial data sets of geological parameters (n=13) were from the Consortium for Spatial Information, Technical University of Denmark, University of East Anglia and the FAO, while the soil parameters (n=42) were from the Harmonized World Soil Database. The aforementioned parameters were regressed to categorical groundwater arsenic concentrations below or above three thresholds: 5 μg/L, 10 μg/L and 50 μg/L to identify respective controlling factors. Boosted regression tree method outperformed logistic regression methods in all three threshold levels in terms of accuracy, specificity and sensitivity, resulting in an improvement of spatial distribution map of probability of groundwater arsenic exceeding all three thresholds when compared to disjunctive-kriging interpolated spatial arsenic map using the same groundwater arsenic dataset. Boosted regression tree models also show that the most important controlling factors of groundwater arsenic distribution include groundwater iron content and well depth for all three thresholds. The probability of a well with iron content higher than 5mg/L to contain greater than 5 μg/L, 10 μg/L and 50 μg/L As is estimated to be more than 91%, 85% and 51%, respectively, while the probability of a well from depth more than 160m to contain more than 5 μg/L, 10 μg/L and 50 μg/L As is estimated to be less than 38%, 25% and 14%, respectively.

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.

A Fast Gradient Method for Nonnegative Sparse Regression With Self-Dictionary

NASA Astrophysics Data System (ADS)

Gillis, Nicolas; Luce, Robert

2018-01-01

A nonnegative matrix factorization (NMF) can be computed efficiently under the separability assumption, which asserts that all the columns of the given input data matrix belong to the cone generated by a (small) subset of them. The provably most robust methods to identify these conic basis columns are based on nonnegative sparse regression and self dictionaries, and require the solution of large-scale convex optimization problems. In this paper we study a particular nonnegative sparse regression model with self dictionary. As opposed to previously proposed models, this model yields a smooth optimization problem where the sparsity is enforced through linear constraints. We show that the Euclidean projection on the polyhedron defined by these constraints can be computed efficiently, and propose a fast gradient method to solve our model. We compare our algorithm with several state-of-the-art methods on synthetic data sets and real-world hyperspectral images.

Bivariate least squares linear regression: Towards a unified analytic formalism. I. Functional models

NASA Astrophysics Data System (ADS)

Caimmi, R.

2011-08-01

Concerning bivariate least squares linear regression, the classical approach pursued for functional models in earlier attempts ( York, 1966, 1969) is reviewed using a new formalism in terms of deviation (matrix) traces which, for unweighted data, reduce to usual quantities leaving aside an unessential (but dimensional) multiplicative factor. Within the framework of classical error models, the dependent variable relates to the independent variable according to the usual additive model. The classes of linear models considered are regression lines in the general case of correlated errors in X and in Y for weighted data, and in the opposite limiting situations of (i) uncorrelated errors in X and in Y, and (ii) completely correlated errors in X and in Y. The special case of (C) generalized orthogonal regression is considered in detail together with well known subcases, namely: (Y) errors in X negligible (ideally null) with respect to errors in Y; (X) errors in Y negligible (ideally null) with respect to errors in X; (O) genuine orthogonal regression; (R) reduced major-axis regression. In the limit of unweighted data, the results determined for functional models are compared with their counterparts related to extreme structural models i.e. the instrumental scatter is negligible (ideally null) with respect to the intrinsic scatter ( Isobe et al., 1990; Feigelson and Babu, 1992). While regression line slope and intercept estimators for functional and structural models necessarily coincide, the contrary holds for related variance estimators even if the residuals obey a Gaussian distribution, with the exception of Y models. An example of astronomical application is considered, concerning the [O/H]-[Fe/H] empirical relations deduced from five samples related to different stars and/or different methods of oxygen abundance determination. For selected samples and assigned methods, different regression models yield consistent results within the errors (∓ σ) for both heteroscedastic and homoscedastic data. Conversely, samples related to different methods produce discrepant results, due to the presence of (still undetected) systematic errors, which implies no definitive statement can be made at present. A comparison is also made between different expressions of regression line slope and intercept variance estimators, where fractional discrepancies are found to be not exceeding a few percent, which grows up to about 20% in the presence of large dispersion data. An extension of the formalism to structural models is left to a forthcoming paper.

Computationally efficient algorithm for Gaussian Process regression in case of structured samples

NASA Astrophysics Data System (ADS)

Belyaev, M.; Burnaev, E.; Kapushev, Y.

2016-04-01

Surrogate modeling is widely used in many engineering problems. Data sets often have Cartesian product structure (for instance factorial design of experiments with missing points). In such case the size of the data set can be very large. Therefore, one of the most popular algorithms for approximation-Gaussian Process regression-can be hardly applied due to its computational complexity. In this paper a computationally efficient approach for constructing Gaussian Process regression in case of data sets with Cartesian product structure is presented. Efficiency is achieved by using a special structure of the data set and operations with tensors. Proposed algorithm has low computational as well as memory complexity compared to existing algorithms. In this work we also introduce a regularization procedure allowing to take into account anisotropy of the data set and avoid degeneracy of regression model.

Influence diagnostics in meta-regression model.

PubMed

Shi, Lei; Zuo, ShanShan; Yu, Dalei; Zhou, Xiaohua

2017-09-01

This paper studies the influence diagnostics in meta-regression model including case deletion diagnostic and local influence analysis. We derive the subset deletion formulae for the estimation of regression coefficient and heterogeneity variance and obtain the corresponding influence measures. The DerSimonian and Laird estimation and maximum likelihood estimation methods in meta-regression are considered, respectively, to derive the results. Internal and external residual and leverage measure are defined. The local influence analysis based on case-weights perturbation scheme, responses perturbation scheme, covariate perturbation scheme, and within-variance perturbation scheme are explored. We introduce a method by simultaneous perturbing responses, covariate, and within-variance to obtain the local influence measure, which has an advantage of capable to compare the influence magnitude of influential studies from different perturbations. An example is used to illustrate the proposed methodology. Copyright © 2017 John Wiley & Sons, Ltd.

Role of regression analysis and variation of rheological data in calculation of pressure drop for sludge pipelines.

PubMed

Farno, E; Coventry, K; Slatter, P; Eshtiaghi, N

2018-06-15

Sludge pumps in wastewater treatment plants are often oversized due to uncertainty in calculation of pressure drop. This issue costs millions of dollars for industry to purchase and operate the oversized pumps. Besides costs, higher electricity consumption is associated with extra CO 2 emission which creates huge environmental impacts. Calculation of pressure drop via current pipe flow theory requires model estimation of flow curve data which depends on regression analysis and also varies with natural variation of rheological data. This study investigates impact of variation of rheological data and regression analysis on variation of pressure drop calculated via current pipe flow theories. Results compare the variation of calculated pressure drop between different models and regression methods and suggest on the suitability of each method. Copyright © 2018 Elsevier Ltd. All rights reserved.

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

Butler, W.J.; Kalasinski, L.A.

In this paper, a generalized logistic regression model for correlated observations is used to analyze epidemiologic data on the frequency of spontaneous abortion among a group of women office workers. The results are compared to those obtained from the use of the standard logistic regression model that assumes statistical independence among all the pregnancies contributed by one woman. In this example, the correlation among pregnancies from the same woman is fairly small and did not have a substantial impact on the magnitude of estimates of parameters of the model. This is due at least partly to the small average numbermore » of pregnancies contributed by each woman.« less

Least-Squares Regression and Spectral Residual Augmented Classical Least-Squares Chemometric Models for Stability-Indicating Analysis of Agomelatine and Its Degradation Products: A Comparative Study.

PubMed

Naguib, Ibrahim A; Abdelrahman, Maha M; El Ghobashy, Mohamed R; Ali, Nesma A

2016-01-01

Two accurate, sensitive, and selective stability-indicating methods are developed and validated for simultaneous quantitative determination of agomelatine (AGM) and its forced degradation products (Deg I and Deg II), whether in pure forms or in pharmaceutical formulations. Partial least-squares regression (PLSR) and spectral residual augmented classical least-squares (SRACLS) are two chemometric models that are being subjected to a comparative study through handling UV spectral data in range (215-350 nm). For proper analysis, a three-factor, four-level experimental design was established, resulting in a training set consisting of 16 mixtures containing different ratios of interfering species. An independent test set consisting of eight mixtures was used to validate the prediction ability of the suggested models. The results presented indicate the ability of mentioned multivariate calibration models to analyze AGM, Deg I, and Deg II with high selectivity and accuracy. The analysis results of the pharmaceutical formulations were statistically compared to the reference HPLC method, with no significant differences observed regarding accuracy and precision. The SRACLS model gives comparable results to the PLSR model; however, it keeps the qualitative spectral information of the classical least-squares algorithm for analyzed components.

Using exploratory regression to identify optimal driving factors for cellular automaton modeling of land use change.

PubMed

Feng, Yongjiu; Tong, Xiaohua

2017-09-22

Defining transition rules is an important issue in cellular automaton (CA)-based land use modeling because these models incorporate highly correlated driving factors. Multicollinearity among correlated driving factors may produce negative effects that must be eliminated from the modeling. Using exploratory regression under pre-defined criteria, we identified all possible combinations of factors from the candidate factors affecting land use change. Three combinations that incorporate five driving factors meeting pre-defined criteria were assessed. With the selected combinations of factors, three logistic regression-based CA models were built to simulate dynamic land use change in Shanghai, China, from 2000 to 2015. For comparative purposes, a CA model with all candidate factors was also applied to simulate the land use change. Simulations using three CA models with multicollinearity eliminated performed better (with accuracy improvements about 3.6%) than the model incorporating all candidate factors. Our results showed that not all candidate factors are necessary for accurate CA modeling and the simulations were not sensitive to changes in statistically non-significant driving factors. We conclude that exploratory regression is an effective method to search for the optimal combinations of driving factors, leading to better land use change models that are devoid of multicollinearity. We suggest identification of dominant factors and elimination of multicollinearity before building land change models, making it possible to simulate more realistic outcomes.

Product unit neural network models for predicting the growth limits of Listeria monocytogenes.

PubMed

Valero, A; Hervás, C; García-Gimeno, R M; Zurera, G

2007-08-01

A new approach to predict the growth/no growth interface of Listeria monocytogenes as a function of storage temperature, pH, citric acid (CA) and ascorbic acid (AA) is presented. A linear logistic regression procedure was performed and a non-linear model was obtained by adding new variables by means of a Neural Network model based on Product Units (PUNN). The classification efficiency of the training data set and the generalization data of the new Logistic Regression PUNN model (LRPU) were compared with Linear Logistic Regression (LLR) and Polynomial Logistic Regression (PLR) models. 92% of the total cases from the LRPU model were correctly classified, an improvement on the percentage obtained using the PLR model (90%) and significantly higher than the results obtained with the LLR model, 80%. On the other hand predictions of LRPU were closer to data observed which permits to design proper formulations in minimally processed foods. This novel methodology can be applied to predictive microbiology for describing growth/no growth interface of food-borne microorganisms such as L. monocytogenes. The optimal balance is trying to find models with an acceptable interpretation capacity and with good ability to fit the data on the boundaries of variable range. The results obtained conclude that these kinds of models might well be very a valuable tool for mathematical modeling.

Two models for identification and predicting behaviour of an induction motor system

NASA Astrophysics Data System (ADS)

Kuo, Chien-Hsun

2018-01-01

System identification or modelling is the process of building mathematical models of dynamical systems based on the available input and output data from the systems. This paper introduces system identification by using ARX (Auto Regressive with eXogeneous input) and ARMAX (Auto Regressive Moving Average with eXogeneous input) models. Through the identified system model, the predicted output could be compared with the measured one to help prevent the motor faults from developing into a catastrophic machine failure and avoid unnecessary costs and delays caused by the need to carry out unscheduled repairs. The induction motor system is illustrated as an example. Numerical and experimental results are shown for the identified induction motor system.

Breast density and parenchymal texture measures as potential risk factors for estrogen-receptor positive breast cancer

NASA Astrophysics Data System (ADS)

Keller, Brad M.; Chen, Jinbo; Conant, Emily F.; Kontos, Despina

2014-03-01

Accurate assessment of a woman's risk to develop specific subtypes of breast cancer is critical for appropriate utilization of chemopreventative measures, such as with tamoxifen in preventing estrogen-receptor positive breast cancer. In this context, we investigate quantitative measures of breast density and parenchymal texture, measures of glandular tissue content and tissue structure, as risk factors for estrogen-receptor positive (ER+) breast cancer. Mediolateral oblique (MLO) view digital mammograms of the contralateral breast from 106 women with unilateral invasive breast cancer were retrospectively analyzed. Breast density and parenchymal texture were analyzed via fully-automated software. Logistic regression with feature selection and was performed to predict ER+ versus ER- cancer status. A combined model considering all imaging measures extracted was compared to baseline models consisting of density-alone and texture-alone features. Area under the curve (AUC) of the receiver operating characteristic (ROC) and Delong's test were used to compare the models' discriminatory capacity for receptor status. The density-alone model had a discriminatory capacity of 0.62 AUC (p=0.05). The texture-alone model had a higher discriminatory capacity of 0.70 AUC (p=0.001), which was not significantly different compared to the density-alone model (p=0.37). In contrast the combined density-texture logistic regression model had a discriminatory capacity of 0.82 AUC (p<0.001), which was statistically significantly higher than both the density-alone (p<0.001) and texture-alone regression models (p=0.04). The combination of breast density and texture measures may have the potential to identify women specifically at risk for estrogen-receptor positive breast cancer and could be useful in triaging women into appropriate risk-reduction strategies.

Development of a drought forecasting model for the Asia-Pacific region using remote sensing and climate data: Focusing on Indonesia

NASA Astrophysics Data System (ADS)

Rhee, Jinyoung; Kim, Gayoung; Im, Jungho

2017-04-01

Three regions of Indonesia with different rainfall characteristics were chosen to develop drought forecast models based on machine learning. The 6-month Standardized Precipitation Index (SPI6) was selected as the target variable. The models' forecast skill was compared to the skill of long-range climate forecast models in terms of drought accuracy and regression mean absolute error (MAE). Indonesian droughts are known to be related to El Nino Southern Oscillation (ENSO) variability despite of regional differences as well as monsoon, local sea surface temperature (SST), other large-scale atmosphere-ocean interactions such as Indian Ocean Dipole (IOD) and Southern Pacific Convergence Zone (SPCZ), and local factors including topography and elevation. Machine learning models are thus to enhance drought forecast skill by combining local and remote SST and remote sensing information reflecting initial drought conditions to the long-range climate forecast model results. A total of 126 machine learning models were developed for the three regions of West Java (JB), West Sumatra (SB), and Gorontalo (GO) and six long-range climate forecast models of MSC_CanCM3, MSC_CanCM4, NCEP, NASA, PNU, POAMA as well as one climatology model based on remote sensing precipitation data, and 1 to 6-month lead times. When compared the results between the machine learning models and the long-range climate forecast models, West Java and Gorontalo regions showed similar characteristics in terms of drought accuracy. Drought accuracy of the long-range climate forecast models were generally higher than the machine learning models with short lead times but the opposite appeared for longer lead times. For West Sumatra, however, the machine learning models and the long-range climate forecast models showed similar drought accuracy. The machine learning models showed smaller regression errors for all three regions especially with longer lead times. Among the three regions, the machine learning models developed for Gorontalo showed the highest drought accuracy and the lowest regression error. West Java showed higher drought accuracy compared to West Sumatra, while West Sumatra showed lower regression error compared to West Java. The lower error in West Sumatra may be because of the smaller sample size used for training and evaluation for the region. Regional differences of forecast skill are determined by the effect of ENSO and the following forecast skill of the long-range climate forecast models. While shown somewhat high in West Sumatra, relative importance of remote sensing variables was mostly low in most cases. High importance of the variables based on long-range climate forecast models indicates that the forecast skill of the machine learning models are mostly determined by the forecast skill of the climate models.

Importance of spatial autocorrelation in modeling bird distributions at a continental scale

USGS Publications Warehouse

Bahn, V.; O'Connor, R.J.; Krohn, W.B.

2006-01-01

Spatial autocorrelation in species' distributions has been recognized as inflating the probability of a type I error in hypotheses tests, causing biases in variable selection, and violating the assumption of independence of error terms in models such as correlation or regression. However, it remains unclear whether these problems occur at all spatial resolutions and extents, and under which conditions spatially explicit modeling techniques are superior. Our goal was to determine whether spatial models were superior at large extents and across many different species. In addition, we investigated the importance of purely spatial effects in distribution patterns relative to the variation that could be explained through environmental conditions. We studied distribution patterns of 108 bird species in the conterminous United States using ten years of data from the Breeding Bird Survey. We compared the performance of spatially explicit regression models with non-spatial regression models using Akaike's information criterion. In addition, we partitioned the variance in species distributions into an environmental, a pure spatial and a shared component. The spatially-explicit conditional autoregressive regression models strongly outperformed the ordinary least squares regression models. In addition, partialling out the spatial component underlying the species' distributions showed that an average of 17% of the explained variation could be attributed to purely spatial effects independent of the spatial autocorrelation induced by the underlying environmental variables. We concluded that location in the range and neighborhood play an important role in the distribution of species. Spatially explicit models are expected to yield better predictions especially for mobile species such as birds, even in coarse-grained models with a large extent. ?? Ecography.

A Poisson regression approach to model monthly hail occurrence in Northern Switzerland using large-scale environmental variables

NASA Astrophysics Data System (ADS)

Madonna, Erica; Ginsbourger, David; Martius, Olivia

2018-05-01

In Switzerland, hail regularly causes substantial damage to agriculture, cars and infrastructure, however, little is known about its long-term variability. To study the variability, the monthly number of days with hail in northern Switzerland is modeled in a regression framework using large-scale predictors derived from ERA-Interim reanalysis. The model is developed and verified using radar-based hail observations for the extended summer season (April-September) in the period 2002-2014. The seasonality of hail is explicitly modeled with a categorical predictor (month) and monthly anomalies of several large-scale predictors are used to capture the year-to-year variability. Several regression models are applied and their performance tested with respect to standard scores and cross-validation. The chosen model includes four predictors: the monthly anomaly of the two meter temperature, the monthly anomaly of the logarithm of the convective available potential energy (CAPE), the monthly anomaly of the wind shear and the month. This model well captures the intra-annual variability and slightly underestimates its inter-annual variability. The regression model is applied to the reanalysis data back in time to 1980. The resulting hail day time series shows an increase of the number of hail days per month, which is (in the model) related to an increase in temperature and CAPE. The trend corresponds to approximately 0.5 days per month per decade. The results of the regression model have been compared to two independent data sets. All data sets agree on the sign of the trend, but the trend is weaker in the other data sets.

Prediction of siRNA potency using sparse logistic regression.

PubMed

Hu, Wei; Hu, John

2014-06-01

RNA interference (RNAi) can modulate gene expression at post-transcriptional as well as transcriptional levels. Short interfering RNA (siRNA) serves as a trigger for the RNAi gene inhibition mechanism, and therefore is a crucial intermediate step in RNAi. There have been extensive studies to identify the sequence characteristics of potent siRNAs. One such study built a linear model using LASSO (Least Absolute Shrinkage and Selection Operator) to measure the contribution of each siRNA sequence feature. This model is simple and interpretable, but it requires a large number of nonzero weights. We have introduced a novel technique, sparse logistic regression, to build a linear model using single-position specific nucleotide compositions which has the same prediction accuracy of the linear model based on LASSO. The weights in our new model share the same general trend as those in the previous model, but have only 25 nonzero weights out of a total 84 weights, a 54% reduction compared to the previous model. Contrary to the linear model based on LASSO, our model suggests that only a few positions are influential on the efficacy of the siRNA, which are the 5' and 3' ends and the seed region of siRNA sequences. We also employed sparse logistic regression to build a linear model using dual-position specific nucleotide compositions, a task LASSO is not able to accomplish well due to its high dimensional nature. Our results demonstrate the superiority of sparse logistic regression as a technique for both feature selection and regression over LASSO in the context of siRNA design.

Cox regression analysis with missing covariates via nonparametric multiple imputation.

PubMed

Hsu, Chiu-Hsieh; Yu, Mandi

2018-01-01

We consider the situation of estimating Cox regression in which some covariates are subject to missing, and there exists additional information (including observed event time, censoring indicator and fully observed covariates) which may be predictive of the missing covariates. We propose to use two working regression models: one for predicting the missing covariates and the other for predicting the missing probabilities. For each missing covariate observation, these two working models are used to define a nearest neighbor imputing set. This set is then used to non-parametrically impute covariate values for the missing observation. Upon the completion of imputation, Cox regression is performed on the multiply imputed datasets to estimate the regression coefficients. In a simulation study, we compare the nonparametric multiple imputation approach with the augmented inverse probability weighted (AIPW) method, which directly incorporates the two working models into estimation of Cox regression, and the predictive mean matching imputation (PMM) method. We show that all approaches can reduce bias due to non-ignorable missing mechanism. The proposed nonparametric imputation method is robust to mis-specification of either one of the two working models and robust to mis-specification of the link function of the two working models. In contrast, the PMM method is sensitive to misspecification of the covariates included in imputation. The AIPW method is sensitive to the selection probability. We apply the approaches to a breast cancer dataset from Surveillance, Epidemiology and End Results (SEER) Program.

Understanding bias in relationships between the food environment and diet quality: the Coronary Artery Risk Development in Young Adults (CARDIA) study.

PubMed

Rummo, Pasquale E; Guilkey, David K; Ng, Shu Wen; Meyer, Katie A; Popkin, Barry M; Reis, Jared P; Shikany, James M; Gordon-Larsen, Penny

2017-12-01

The relationship between food environment exposures and diet behaviours is unclear, possibly because the majority of studies ignore potential residual confounding. We used 20 years (1985-1986, 1992-1993 2005-2006) of data from the Coronary Artery Risk Development in Young Adults (CARDIA) study across four US cities (Birmingham, Alabama; Chicago, Illinois; Minneapolis, Minnesota; Oakland, California) and instrumental variables (IV) regression to obtain causal estimates of longitudinal associations between the percentage of neighbourhood food outlets (per total food outlets within 1 km network distance of respondent residence) and an a priori diet quality score, with higher scores indicating higher diet quality. To assess the presence and magnitude of bias related to residual confounding, we compared results from causal models (IV regression) to non-causal models, including ordinary least squares regression, which does not account for residual confounding at all and fixed-effects regression, which only controls for time-invariant unmeasured characteristics. The mean diet quality score across follow-up was 63.4 (SD=12.7). A 10% increase in fast food restaurants (relative to full-service restaurants) was associated with a lower diet quality score over time using IV regression (β=-1.01, 95% CI -1.99 to -0.04); estimates were attenuated using non-causal models. The percentage of neighbourhood convenience and grocery stores (relative to supermarkets) was not associated with diet quality in any model, but estimates from non-causal models were similarly attenuated compared with causal models. Ignoring residual confounding may generate biased estimated effects of neighbourhood food outlets on diet outcomes and may have contributed to weak findings in the food environment literature. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2017. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

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.

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.

Use of genetic programming, logistic regression, and artificial neural nets to predict readmission after coronary artery bypass surgery.

PubMed

Engoren, Milo; Habib, Robert H; Dooner, John J; Schwann, Thomas A

2013-08-01

As many as 14 % of patients undergoing coronary artery bypass surgery are readmitted within 30 days. Readmission is usually the result of morbidity and may lead to death. The purpose of this study is to develop and compare statistical and genetic programming models to predict readmission. Patients were divided into separate Construction and Validation populations. Using 88 variables, logistic regression, genetic programs, and artificial neural nets were used to develop predictive models. Models were first constructed and tested on the Construction populations, then validated on the Validation population. Areas under the receiver operator characteristic curves (AU ROC) were used to compare the models. Two hundred and two patients (7.6 %) in the 2,644 patient Construction group and 216 (8.0 %) of the 2,711 patient Validation group were re-admitted within 30 days of CABG surgery. Logistic regression predicted readmission with AU ROC = .675 ± .021 in the Construction group. Genetic programs significantly improved the accuracy, AU ROC = .767 ± .001, p < .001). Artificial neural nets were less accurate with AU ROC = 0.597 ± .001 in the Construction group. Predictive accuracy of all three techniques fell in the Validation group. However, the accuracy of genetic programming (AU ROC = .654 ± .001) was still trivially but statistically non-significantly better than that of the logistic regression (AU ROC = .644 ± .020, p = .61). Genetic programming and logistic regression provide alternative methods to predict readmission that are similarly accurate.

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.

Groundwater-level prediction using multiple linear regression and artificial neural network techniques: a comparative assessment

NASA Astrophysics Data System (ADS)

Sahoo, Sasmita; Jha, Madan K.

2013-12-01

The potential of multiple linear regression (MLR) and artificial neural network (ANN) techniques in predicting transient water levels over a groundwater basin were compared. MLR and ANN modeling was carried out at 17 sites in Japan, considering all significant inputs: rainfall, ambient temperature, river stage, 11 seasonal dummy variables, and influential lags of rainfall, ambient temperature, river stage and groundwater level. Seventeen site-specific ANN models were developed, using multi-layer feed-forward neural networks trained with Levenberg-Marquardt backpropagation algorithms. The performance of the models was evaluated using statistical and graphical indicators. Comparison of the goodness-of-fit statistics of the MLR models with those of the ANN models indicated that there is better agreement between the ANN-predicted groundwater levels and the observed groundwater levels at all the sites, compared to the MLR. This finding was supported by the graphical indicators and the residual analysis. Thus, it is concluded that the ANN technique is superior to the MLR technique in predicting spatio-temporal distribution of groundwater levels in a basin. However, considering the practical advantages of the MLR technique, it is recommended as an alternative and cost-effective groundwater modeling tool.

Bayesian logistic regression approaches to predict incorrect DRG assignment.

PubMed

Suleiman, Mani; Demirhan, Haydar; Boyd, Leanne; Girosi, Federico; Aksakalli, Vural

2018-05-07

Episodes of care involving similar diagnoses and treatments and requiring similar levels of resource utilisation are grouped to the same Diagnosis-Related Group (DRG). In jurisdictions which implement DRG based payment systems, DRGs are a major determinant of funding for inpatient care. Hence, service providers often dedicate auditing staff to the task of checking that episodes have been coded to the correct DRG. The use of statistical models to estimate an episode's probability of DRG error can significantly improve the efficiency of clinical coding audits. This study implements Bayesian logistic regression models with weakly informative prior distributions to estimate the likelihood that episodes require a DRG revision, comparing these models with each other and to classical maximum likelihood estimates. All Bayesian approaches had more stable model parameters than maximum likelihood. The best performing Bayesian model improved overall classification per- formance by 6% compared to maximum likelihood, with a 34% gain compared to random classification, respectively. We found that the original DRG, coder and the day of coding all have a significant effect on the likelihood of DRG error. Use of Bayesian approaches has improved model parameter stability and classification accuracy. This method has already lead to improved audit efficiency in an operational capacity.

The PIT-trap-A "model-free" bootstrap procedure for inference about regression models with discrete, multivariate responses.

PubMed

Warton, David I; Thibaut, Loïc; Wang, Yi Alice

2017-01-01

Bootstrap methods are widely used in statistics, and bootstrapping of residuals can be especially useful in the regression context. However, difficulties are encountered extending residual resampling to regression settings where residuals are not identically distributed (thus not amenable to bootstrapping)-common examples including logistic or Poisson regression and generalizations to handle clustered or multivariate data, such as generalised estimating equations. We propose a bootstrap method based on probability integral transform (PIT-) residuals, which we call the PIT-trap, which assumes data come from some marginal distribution F of known parametric form. This method can be understood as a type of "model-free bootstrap", adapted to the problem of discrete and highly multivariate data. PIT-residuals have the key property that they are (asymptotically) pivotal. The PIT-trap thus inherits the key property, not afforded by any other residual resampling approach, that the marginal distribution of data can be preserved under PIT-trapping. This in turn enables the derivation of some standard bootstrap properties, including second-order correctness of pivotal PIT-trap test statistics. In multivariate data, bootstrapping rows of PIT-residuals affords the property that it preserves correlation in data without the need for it to be modelled, a key point of difference as compared to a parametric bootstrap. The proposed method is illustrated on an example involving multivariate abundance data in ecology, and demonstrated via simulation to have improved properties as compared to competing resampling methods.

The PIT-trap—A “model-free” bootstrap procedure for inference about regression models with discrete, multivariate responses

PubMed Central

Thibaut, Loïc; Wang, Yi Alice

2017-01-01

Bootstrap methods are widely used in statistics, and bootstrapping of residuals can be especially useful in the regression context. However, difficulties are encountered extending residual resampling to regression settings where residuals are not identically distributed (thus not amenable to bootstrapping)—common examples including logistic or Poisson regression and generalizations to handle clustered or multivariate data, such as generalised estimating equations. We propose a bootstrap method based on probability integral transform (PIT-) residuals, which we call the PIT-trap, which assumes data come from some marginal distribution F of known parametric form. This method can be understood as a type of “model-free bootstrap”, adapted to the problem of discrete and highly multivariate data. PIT-residuals have the key property that they are (asymptotically) pivotal. The PIT-trap thus inherits the key property, not afforded by any other residual resampling approach, that the marginal distribution of data can be preserved under PIT-trapping. This in turn enables the derivation of some standard bootstrap properties, including second-order correctness of pivotal PIT-trap test statistics. In multivariate data, bootstrapping rows of PIT-residuals affords the property that it preserves correlation in data without the need for it to be modelled, a key point of difference as compared to a parametric bootstrap. The proposed method is illustrated on an example involving multivariate abundance data in ecology, and demonstrated via simulation to have improved properties as compared to competing resampling methods. PMID:28738071

Three methods to construct predictive models using logistic regression and likelihood ratios to facilitate adjustment for pretest probability give similar results.

PubMed

Chan, Siew Foong; Deeks, Jonathan J; Macaskill, Petra; Irwig, Les

2008-01-01

To compare three predictive models based on logistic regression to estimate adjusted likelihood ratios allowing for interdependency between diagnostic variables (tests). This study was a review of the theoretical basis, assumptions, and limitations of published models; and a statistical extension of methods and application to a case study of the diagnosis of obstructive airways disease based on history and clinical examination. Albert's method includes an offset term to estimate an adjusted likelihood ratio for combinations of tests. Spiegelhalter and Knill-Jones method uses the unadjusted likelihood ratio for each test as a predictor and computes shrinkage factors to allow for interdependence. Knottnerus' method differs from the other methods because it requires sequencing of tests, which limits its application to situations where there are few tests and substantial data. Although parameter estimates differed between the models, predicted "posttest" probabilities were generally similar. Construction of predictive models using logistic regression is preferred to the independence Bayes' approach when it is important to adjust for dependency of tests errors. Methods to estimate adjusted likelihood ratios from predictive models should be considered in preference to a standard logistic regression model to facilitate ease of interpretation and application. Albert's method provides the most straightforward approach.

Aggregating the response in time series regression models, applied to weather-related cardiovascular mortality

NASA Astrophysics Data System (ADS)

Masselot, Pierre; Chebana, Fateh; Bélanger, Diane; St-Hilaire, André; Abdous, Belkacem; Gosselin, Pierre; Ouarda, Taha B. M. J.

2018-07-01

In environmental epidemiology studies, health response data (e.g. hospitalization or mortality) are often noisy because of hospital organization and other social factors. The noise in the data can hide the true signal related to the exposure. The signal can be unveiled by performing a temporal aggregation on health data and then using it as the response in regression analysis. From aggregated series, a general methodology is introduced to account for the particularities of an aggregated response in a regression setting. This methodology can be used with usually applied regression models in weather-related health studies, such as generalized additive models (GAM) and distributed lag nonlinear models (DLNM). In particular, the residuals are modelled using an autoregressive-moving average (ARMA) model to account for the temporal dependence. The proposed methodology is illustrated by modelling the influence of temperature on cardiovascular mortality in Canada. A comparison with classical DLNMs is provided and several aggregation methods are compared. Results show that there is an increase in the fit quality when the response is aggregated, and that the estimated relationship focuses more on the outcome over several days than the classical DLNM. More precisely, among various investigated aggregation schemes, it was found that an aggregation with an asymmetric Epanechnikov kernel is more suited for studying the temperature-mortality relationship.

Determination of suitable drying curve model for bread moisture loss during baking

NASA Astrophysics Data System (ADS)

Soleimani Pour-Damanab, A. R.; Jafary, A.; Rafiee, S.

2013-03-01

This study presents mathematical modelling of bread moisture loss or drying during baking in a conventional bread baking process. In order to estimate and select the appropriate moisture loss curve equation, 11 different models, semi-theoretical and empirical, were applied to the experimental data and compared according to their correlation coefficients, chi-squared test and root mean square error which were predicted by nonlinear regression analysis. Consequently, of all the drying models, a Page model was selected as the best one, according to the correlation coefficients, chi-squared test, and root mean square error values and its simplicity. Mean absolute estimation error of the proposed model by linear regression analysis for natural and forced convection modes was 2.43, 4.74%, respectively.

Multiple-Shrinkage Multinomial Probit Models with Applications to Simulating Geographies in Public Use Data.

PubMed

Burgette, Lane F; Reiter, Jerome P

2013-06-01

Multinomial outcomes with many levels can be challenging to model. Information typically accrues slowly with increasing sample size, yet the parameter space expands rapidly with additional covariates. Shrinking all regression parameters towards zero, as often done in models of continuous or binary response variables, is unsatisfactory, since setting parameters equal to zero in multinomial models does not necessarily imply "no effect." We propose an approach to modeling multinomial outcomes with many levels based on a Bayesian multinomial probit (MNP) model and a multiple shrinkage prior distribution for the regression parameters. The prior distribution encourages the MNP regression parameters to shrink toward a number of learned locations, thereby substantially reducing the dimension of the parameter space. Using simulated data, we compare the predictive performance of this model against two other recently-proposed methods for big multinomial models. The results suggest that the fully Bayesian, multiple shrinkage approach can outperform these other methods. We apply the multiple shrinkage MNP to simulating replacement values for areal identifiers, e.g., census tract indicators, in order to protect data confidentiality in public use datasets.

Statistical downscaling of precipitation using long short-term memory recurrent neural networks

NASA Astrophysics Data System (ADS)

Misra, Saptarshi; Sarkar, Sudeshna; Mitra, Pabitra

2017-11-01

Hydrological impacts of global climate change on regional scale are generally assessed by downscaling large-scale climatic variables, simulated by General Circulation Models (GCMs), to regional, small-scale hydrometeorological variables like precipitation, temperature, etc. In this study, we propose a new statistical downscaling model based on Recurrent Neural Network with Long Short-Term Memory which captures the spatio-temporal dependencies in local rainfall. The previous studies have used several other methods such as linear regression, quantile regression, kernel regression, beta regression, and artificial neural networks. Deep neural networks and recurrent neural networks have been shown to be highly promising in modeling complex and highly non-linear relationships between input and output variables in different domains and hence we investigated their performance in the task of statistical downscaling. We have tested this model on two datasets—one on precipitation in Mahanadi basin in India and the second on precipitation in Campbell River basin in Canada. Our autoencoder coupled long short-term memory recurrent neural network model performs the best compared to other existing methods on both the datasets with respect to temporal cross-correlation, mean squared error, and capturing the extremes.

Quality of life in breast cancer patients--a quantile regression analysis.

PubMed

Pourhoseingholi, Mohamad Amin; Safaee, Azadeh; Moghimi-Dehkordi, Bijan; Zeighami, Bahram; Faghihzadeh, Soghrat; Tabatabaee, Hamid Reza; Pourhoseingholi, Asma

2008-01-01

Quality of life study has an important role in health care especially in chronic diseases, in clinical judgment and in medical resources supplying. Statistical tools like linear regression are widely used to assess the predictors of quality of life. But when the response is not normal the results are misleading. The aim of this study is to determine the predictors of quality of life in breast cancer patients, using quantile regression model and compare to linear regression. A cross-sectional study conducted on 119 breast cancer patients that admitted and treated in chemotherapy ward of Namazi hospital in Shiraz. We used QLQ-C30 questionnaire to assessment quality of life in these patients. A quantile regression was employed to assess the assocciated factors and the results were compared to linear regression. All analysis carried out using SAS. The mean score for the global health status for breast cancer patients was 64.92+/-11.42. Linear regression showed that only grade of tumor, occupational status, menopausal status, financial difficulties and dyspnea were statistically significant. In spite of linear regression, financial difficulties were not significant in quantile regression analysis and dyspnea was only significant for first quartile. Also emotion functioning and duration of disease statistically predicted the QOL score in the third quartile. The results have demonstrated that using quantile regression leads to better interpretation and richer inference about predictors of the breast cancer patient quality of life.

Bayesian regression analyses of radiation modality effects on pericardial and pleural effusion and survival in esophageal cancer.

PubMed

He, Liru; Chapple, Andrew; Liao, Zhongxing; Komaki, Ritsuko; Thall, Peter F; Lin, Steven H

2016-10-01

To evaluate radiation modality effects on pericardial effusion (PCE), pleural effusion (PE) and survival in esophageal cancer (EC) patients. We analyzed data from 470 EC patients treated with definitive concurrent chemoradiotherapy (CRT). Bayesian semi-competing risks (SCR) regression models were fit to assess effects of radiation modality and prognostic covariates on the risks of PCE and PE, and death either with or without these preceding events. Bayesian piecewise exponential regression models were fit for overall survival, the time to PCE or death, and the time to PE or death. All models included propensity score as a covariate to correct for potential selection bias. Median times to onset of PCE and PE after RT were 7.1 and 6.1months for IMRT, and 6.5 and 5.4months for 3DCRT, respectively. Compared to 3DCRT, the IMRT group had significantly lower risks of PE, PCE, and death. The respective probabilities of a patient being alive without either PCE or PE at 3-years and 5-years were 0.29 and 0.21 for IMRT compared to 0.13 and 0.08 for 3DCRT. In the SCR regression analyses, IMRT was associated with significantly lower risks of PCE (HR=0.26) and PE (HR=0.49), and greater overall survival (probability of beneficial effect (pbe)>0.99), after controlling for known clinical prognostic factors. IMRT reduces the incidence and postpones the onset of PCE and PE, and increases survival probability, compared to 3DCRT. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

Wildfire suppression cost forecasts from the US Forest Service

Treesearch

Karen L. Abt; Jeffrey P. Prestemon; Krista M. Gebert

2009-01-01

The US Forest Service and other land-management agencies seek better tools for nticipating future expenditures for wildfire suppression. We developed regression models for forecasting US Forest Service suppression spending at 1-, 2-, and 3-year lead times. We compared these models to another readily available forecast model, the 10-year moving average model,...

A study of machine learning regression methods for major elemental analysis of rocks using laser-induced breakdown spectroscopy

NASA Astrophysics Data System (ADS)

Boucher, Thomas F.; Ozanne, Marie V.; Carmosino, Marco L.; Dyar, M. Darby; Mahadevan, Sridhar; Breves, Elly A.; Lepore, Kate H.; Clegg, Samuel M.

2015-05-01

The ChemCam instrument on the Mars Curiosity rover is generating thousands of LIBS spectra and bringing interest in this technique to public attention. The key to interpreting Mars or any other types of LIBS data are calibrations that relate laboratory standards to unknowns examined in other settings and enable predictions of chemical composition. Here, LIBS spectral data are analyzed using linear regression methods including partial least squares (PLS-1 and PLS-2), principal component regression (PCR), least absolute shrinkage and selection operator (lasso), elastic net, and linear support vector regression (SVR-Lin). These were compared against results from nonlinear regression methods including kernel principal component regression (K-PCR), polynomial kernel support vector regression (SVR-Py) and k-nearest neighbor (kNN) regression to discern the most effective models for interpreting chemical abundances from LIBS spectra of geological samples. The results were evaluated for 100 samples analyzed with 50 laser pulses at each of five locations averaged together. Wilcoxon signed-rank tests were employed to evaluate the statistical significance of differences among the nine models using their predicted residual sum of squares (PRESS) to make comparisons. For MgO, SiO2, Fe2O3, CaO, and MnO, the sparse models outperform all the others except for linear SVR, while for Na2O, K2O, TiO2, and P2O5, the sparse methods produce inferior results, likely because their emission lines in this energy range have lower transition probabilities. The strong performance of the sparse methods in this study suggests that use of dimensionality-reduction techniques as a preprocessing step may improve the performance of the linear models. Nonlinear methods tend to overfit the data and predict less accurately, while the linear methods proved to be more generalizable with better predictive performance. These results are attributed to the high dimensionality of the data (6144 channels) relative to the small number of samples studied. The best-performing models were SVR-Lin for SiO2, MgO, Fe2O3, and Na2O, lasso for Al2O3, elastic net for MnO, and PLS-1 for CaO, TiO2, and K2O. Although these differences in model performance between methods were identified, most of the models produce comparable results when p ≤ 0.05 and all techniques except kNN produced statistically-indistinguishable results. It is likely that a combination of models could be used together to yield a lower total error of prediction, depending on the requirements of the user.

Regional vertical total electron content (VTEC) modeling together with satellite and receiver differential code biases (DCBs) using semi-parametric multivariate adaptive regression B-splines (SP-BMARS)

NASA Astrophysics Data System (ADS)

Durmaz, Murat; Karslioglu, Mahmut Onur

2015-04-01

There are various global and regional methods that have been proposed for the modeling of ionospheric vertical total electron content (VTEC). Global distribution of VTEC is usually modeled by spherical harmonic expansions, while tensor products of compactly supported univariate B-splines can be used for regional modeling. In these empirical parametric models, the coefficients of the basis functions as well as differential code biases (DCBs) of satellites and receivers can be treated as unknown parameters which can be estimated from geometry-free linear combinations of global positioning system observables. In this work we propose a new semi-parametric multivariate adaptive regression B-splines (SP-BMARS) method for the regional modeling of VTEC together with satellite and receiver DCBs, where the parametric part of the model is related to the DCBs as fixed parameters and the non-parametric part adaptively models the spatio-temporal distribution of VTEC. The latter is based on multivariate adaptive regression B-splines which is a non-parametric modeling technique making use of compactly supported B-spline basis functions that are generated from the observations automatically. This algorithm takes advantage of an adaptive scale-by-scale model building strategy that searches for best-fitting B-splines to the data at each scale. The VTEC maps generated from the proposed method are compared numerically and visually with the global ionosphere maps (GIMs) which are provided by the Center for Orbit Determination in Europe (CODE). The VTEC values from SP-BMARS and CODE GIMs are also compared with VTEC values obtained through calibration using local ionospheric model. The estimated satellite and receiver DCBs from the SP-BMARS model are compared with the CODE distributed DCBs. The results show that the SP-BMARS algorithm can be used to estimate satellite and receiver DCBs while adaptively and flexibly modeling the daily regional VTEC.

Mountain pine beetle attack in ponderosa pine: Comparing methods for rating susceptibility

Treesearch

David C. Chojnacky; Barbara J. Bentz; Jesse A. Logan

2000-01-01

Two empirical methods for rating susceptibility of mountain pine beetle attack in ponderosa pine were evaluated. The methods were compared to stand data modeled to objectively rate each sampled stand for susceptibly to bark-beetle attack. Data on bark-beetle attacks, from a survey of 45 sites throughout the Colorado Plateau, were modeled using logistic regression to...

Conditional Poisson models: a flexible alternative to conditional logistic case cross-over analysis.

PubMed

Armstrong, Ben G; Gasparrini, Antonio; Tobias, Aurelio

2014-11-24

The time stratified case cross-over approach is a popular alternative to conventional time series regression for analysing associations between time series of environmental exposures (air pollution, weather) and counts of health outcomes. These are almost always analyzed using conditional logistic regression on data expanded to case-control (case crossover) format, but this has some limitations. In particular adjusting for overdispersion and auto-correlation in the counts is not possible. It has been established that a Poisson model for counts with stratum indicators gives identical estimates to those from conditional logistic regression and does not have these limitations, but it is little used, probably because of the overheads in estimating many stratum parameters. The conditional Poisson model avoids estimating stratum parameters by conditioning on the total event count in each stratum, thus simplifying the computing and increasing the number of strata for which fitting is feasible compared with the standard unconditional Poisson model. Unlike the conditional logistic model, the conditional Poisson model does not require expanding the data, and can adjust for overdispersion and auto-correlation. It is available in Stata, R, and other packages. By applying to some real data and using simulations, we demonstrate that conditional Poisson models were simpler to code and shorter to run than are conditional logistic analyses and can be fitted to larger data sets than possible with standard Poisson models. Allowing for overdispersion or autocorrelation was possible with the conditional Poisson model but when not required this model gave identical estimates to those from conditional logistic regression. Conditional Poisson regression models provide an alternative to case crossover analysis of stratified time series data with some advantages. The conditional Poisson model can also be used in other contexts in which primary control for confounding is by fine stratification.

On The Impact of Climate Change to Agricultural Productivity in East Java

NASA Astrophysics Data System (ADS)

Kuswanto, Heri; Salamah, Mutiah; Mumpuni Retnaningsih, Sri; Dwi Prastyo, Dedy

2018-03-01

Many researches showed that climate change has significant impact on agricultural sector, which threats the food security especially in developing countries. It has been observed also that the climate change increases the intensity of extreme events. This research investigated the impact climate to the agricultural productivity in East Java, as one of the main rice producers in Indonesia. Standard regression as well as panel regression models have been performed in order to find the best model which is able to describe the climate change impact. The analysis found that the fixed effect model of panel regression outperforms the others showing that climate change had negatively impacted the rice productivity in East Java. The effect in Malang and Pasuruan were almost the same, while the impact in Sumenep was the least one compared to other districts.

Analysis of Binary Adherence Data in the Setting of Polypharmacy: A Comparison of Different Approaches

PubMed Central

Esserman, Denise A.; Moore, Charity G.; Roth, Mary T.

2009-01-01

Older community dwelling adults often take multiple medications for numerous chronic diseases. Non-adherence to these medications can have a large public health impact. Therefore, the measurement and modeling of medication adherence in the setting of polypharmacy is an important area of research. We apply a variety of different modeling techniques (standard linear regression; weighted linear regression; adjusted linear regression; naïve logistic regression; beta-binomial (BB) regression; generalized estimating equations (GEE)) to binary medication adherence data from a study in a North Carolina based population of older adults, where each medication an individual was taking was classified as adherent or non-adherent. In addition, through simulation we compare these different methods based on Type I error rates, bias, power, empirical 95% coverage, and goodness of fit. We find that estimation and inference using GEE is robust to a wide variety of scenarios and we recommend using this in the setting of polypharmacy when adherence is dichotomously measured for multiple medications per person. PMID:20414358

Calibrated Multivariate Regression with Application to Neural Semantic Basis Discovery.

PubMed

Liu, Han; Wang, Lie; Zhao, Tuo

2015-08-01

We propose a calibrated multivariate regression method named CMR for fitting high dimensional multivariate regression models. Compared with existing methods, CMR calibrates regularization for each regression task with respect to its noise level so that it simultaneously attains improved finite-sample performance and tuning insensitiveness. Theoretically, we provide sufficient conditions under which CMR achieves the optimal rate of convergence in parameter estimation. Computationally, we propose an efficient smoothed proximal gradient algorithm with a worst-case numerical rate of convergence O (1/ ϵ ), where ϵ is a pre-specified accuracy of the objective function value. We conduct thorough numerical simulations to illustrate that CMR consistently outperforms other high dimensional multivariate regression methods. We also apply CMR to solve a brain activity prediction problem and find that it is as competitive as a handcrafted model created by human experts. The R package camel implementing the proposed method is available on the Comprehensive R Archive Network http://cran.r-project.org/web/packages/camel/.

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

Remote sensing and GIS-based landslide hazard analysis and cross-validation using multivariate logistic regression model on three test areas in Malaysia

NASA Astrophysics Data System (ADS)

Pradhan, Biswajeet

2010-05-01

This paper presents the results of the cross-validation of a multivariate logistic regression model using remote sensing data and GIS for landslide hazard analysis on the Penang, Cameron, and Selangor areas in Malaysia. Landslide locations in the study areas were identified by interpreting aerial photographs and satellite images, supported by field surveys. SPOT 5 and Landsat TM satellite imagery were used to map landcover and vegetation index, respectively. Maps of topography, soil type, lineaments and land cover were constructed from the spatial datasets. Ten factors which influence landslide occurrence, i.e., slope, aspect, curvature, distance from drainage, lithology, distance from lineaments, soil type, landcover, rainfall precipitation, and normalized difference vegetation index (ndvi), were extracted from the spatial database and the logistic regression coefficient of each factor was computed. Then the landslide hazard was analysed using the multivariate logistic regression coefficients derived not only from the data for the respective area but also using the logistic regression coefficients calculated from each of the other two areas (nine hazard maps in all) as a cross-validation of the model. For verification of the model, the results of the analyses were then compared with the field-verified landslide locations. Among the three cases of the application of logistic regression coefficient in the same study area, the case of Selangor based on the Selangor logistic regression coefficients showed the highest accuracy (94%), where as Penang based on the Penang coefficients showed the lowest accuracy (86%). Similarly, among the six cases from the cross application of logistic regression coefficient in other two areas, the case of Selangor based on logistic coefficient of Cameron showed highest (90%) prediction accuracy where as the case of Penang based on the Selangor logistic regression coefficients showed the lowest accuracy (79%). Qualitatively, the cross application model yields reasonable results which can be used for preliminary landslide hazard mapping.

A Modified Double Multiple Nonlinear Regression Constitutive Equation for Modeling and Prediction of High Temperature Flow Behavior of BFe10-1-2 Alloy

NASA Astrophysics Data System (ADS)

Cai, Jun; Wang, Kuaishe; Shi, Jiamin; Wang, Wen; Liu, Yingying

2018-01-01

Constitutive analysis for hot working of BFe10-1-2 alloy was carried out by using experimental stress-strain data from isothermal hot compression tests, in a wide range of temperature of 1,023 1,273 K, and strain rate range of 0.001 10 s-1. A constitutive equation based on modified double multiple nonlinear regression was proposed considering the independent effects of strain, strain rate, temperature and their interrelation. The predicted flow stress data calculated from the developed equation was compared with the experimental data. Correlation coefficient (R), average absolute relative error (AARE) and relative errors were introduced to verify the validity of the developed constitutive equation. Subsequently, a comparative study was made on the capability of strain-compensated Arrhenius-type constitutive model. The results showed that the developed constitutive equation based on modified double multiple nonlinear regression could predict flow stress of BFe10-1-2 alloy with good correlation and generalization.

The Impact of School Improvement Grants on Achievement: Plans for a National Evaluation Using a Regression Discontinuity Design

ERIC Educational Resources Information Center

Deke, John; Dragoset, Lisa

2015-01-01

Does receipt of School Improvement Grants (SIG) funding to implement a school intervention model have an impact on outcomes for low-performing schools? This study answers this question using a regression discontinuity design (RDD) that exploits cutoff values on the continuous variables used to define SIG eligibility tiers, comparing outcomes in…

A Comparative Study of Classification and Regression Algorithms for Modelling Students' Academic Performance

ERIC Educational Resources Information Center

Strecht, Pedro; Cruz, Luís; Soares, Carlos; Mendes-Moreira, João; Abreu, Rui

2015-01-01

Predicting the success or failure of a student in a course or program is a problem that has recently been addressed using data mining techniques. In this paper we evaluate some of the most popular classification and regression algorithms on this problem. We address two problems: prediction of approval/failure and prediction of grade. The former is…

Comparison of integrated clustering methods for accurate and stable prediction of building energy consumption data

DOE PAGES

Hsu, David

2015-09-27

Clustering methods are often used to model energy consumption for two reasons. First, clustering is often used to process data and to improve the predictive accuracy of subsequent energy models. Second, stable clusters that are reproducible with respect to non-essential changes can be used to group, target, and interpret observed subjects. However, it is well known that clustering methods are highly sensitive to the choice of algorithms and variables. This can lead to misleading assessments of predictive accuracy and mis-interpretation of clusters in policymaking. This paper therefore introduces two methods to the modeling of energy consumption in buildings: clusterwise regression,more » also known as latent class regression, which integrates clustering and regression simultaneously; and cluster validation methods to measure stability. Using a large dataset of multifamily buildings in New York City, clusterwise regression is compared to common two-stage algorithms that use K-means and model-based clustering with linear regression. Predictive accuracy is evaluated using 20-fold cross validation, and the stability of the perturbed clusters is measured using the Jaccard coefficient. These results show that there seems to be an inherent tradeoff between prediction accuracy and cluster stability. This paper concludes by discussing which clustering methods may be appropriate for different analytical purposes.« less

Predicting chemical bioavailability using microarray gene expression data and regression modeling: A tale of three explosive compounds.

PubMed

Gong, Ping; Nan, Xiaofei; Barker, Natalie D; Boyd, Robert E; Chen, Yixin; Wilkins, Dawn E; Johnson, David R; Suedel, Burton C; Perkins, Edward J

2016-03-08

Chemical bioavailability is an important dose metric in environmental risk assessment. Although many approaches have been used to evaluate bioavailability, not a single approach is free from limitations. Previously, we developed a new genomics-based approach that integrated microarray technology and regression modeling for predicting bioavailability (tissue residue) of explosives compounds in exposed earthworms. In the present study, we further compared 18 different regression models and performed variable selection simultaneously with parameter estimation. This refined approach was applied to both previously collected and newly acquired earthworm microarray gene expression datasets for three explosive compounds. Our results demonstrate that a prediction accuracy of R(2) = 0.71-0.82 was achievable at a relatively low model complexity with as few as 3-10 predictor genes per model. These results are much more encouraging than our previous ones. This study has demonstrated that our approach is promising for bioavailability measurement, which warrants further studies of mixed contamination scenarios in field settings.

Regularization Paths for Conditional Logistic Regression: The clogitL1 Package.

PubMed

Reid, Stephen; Tibshirani, Rob

2014-07-01

We apply the cyclic coordinate descent algorithm of Friedman, Hastie, and Tibshirani (2010) to the fitting of a conditional logistic regression model with lasso [Formula: see text] and elastic net penalties. The sequential strong rules of Tibshirani, Bien, Hastie, Friedman, Taylor, Simon, and Tibshirani (2012) are also used in the algorithm and it is shown that these offer a considerable speed up over the standard coordinate descent algorithm with warm starts. Once implemented, the algorithm is used in simulation studies to compare the variable selection and prediction performance of the conditional logistic regression model against that of its unconditional (standard) counterpart. We find that the conditional model performs admirably on datasets drawn from a suitable conditional distribution, outperforming its unconditional counterpart at variable selection. The conditional model is also fit to a small real world dataset, demonstrating how we obtain regularization paths for the parameters of the model and how we apply cross validation for this method where natural unconditional prediction rules are hard to come by.

Regularization Paths for Conditional Logistic Regression: The clogitL1 Package

PubMed Central

Reid, Stephen; Tibshirani, Rob

2014-01-01

We apply the cyclic coordinate descent algorithm of Friedman, Hastie, and Tibshirani (2010) to the fitting of a conditional logistic regression model with lasso (ℓ1) and elastic net penalties. The sequential strong rules of Tibshirani, Bien, Hastie, Friedman, Taylor, Simon, and Tibshirani (2012) are also used in the algorithm and it is shown that these offer a considerable speed up over the standard coordinate descent algorithm with warm starts. Once implemented, the algorithm is used in simulation studies to compare the variable selection and prediction performance of the conditional logistic regression model against that of its unconditional (standard) counterpart. We find that the conditional model performs admirably on datasets drawn from a suitable conditional distribution, outperforming its unconditional counterpart at variable selection. The conditional model is also fit to a small real world dataset, demonstrating how we obtain regularization paths for the parameters of the model and how we apply cross validation for this method where natural unconditional prediction rules are hard to come by. PMID:26257587

Estimating Soil Cation Exchange Capacity from Soil Physical and Chemical Properties

NASA Astrophysics Data System (ADS)

Bateni, S. M.; Emamgholizadeh, S.; Shahsavani, D.

2014-12-01

The soil Cation Exchange Capacity (CEC) is an important soil characteristic that has many applications in soil science and environmental studies. For example, CEC influences soil fertility by controlling the exchange of ions in the soil. Measurement of CEC is costly and difficult. Consequently, several studies attempted to obtain CEC from readily measurable soil physical and chemical properties such as soil pH, organic matter, soil texture, bulk density, and particle size distribution. These studies have often used multiple regression or artificial neural network models. Regression-based models cannot capture the intricate relationship between CEC and soil physical and chemical attributes and provide inaccurate CEC estimates. Although neural network models perform better than regression methods, they act like a black-box and cannot generate an explicit expression for retrieval of CEC from soil properties. In a departure with regression and neural network models, this study uses Genetic Expression Programming (GEP) and Multivariate Adaptive Regression Splines (MARS) to estimate CEC from easily measurable soil variables such as clay, pH, and OM. CEC estimates from GEP and MARS are compared with measurements at two field sites in Iran. Results show that GEP and MARS can estimate CEC accurately. Also, the MARS model performs slightly better than GEP. Finally, a sensitivity test indicates that organic matter and pH have respectively the least and the most significant impact on CEC.

Convergent Time-Varying Regression Models for Data Streams: Tracking Concept Drift by the Recursive Parzen-Based Generalized Regression Neural Networks.

PubMed

Duda, Piotr; Jaworski, Maciej; Rutkowski, Leszek

2018-03-01

One of the greatest challenges in data mining is related to processing and analysis of massive data streams. Contrary to traditional static data mining problems, data streams require that each element is processed only once, the amount of allocated memory is constant and the models incorporate changes of investigated streams. A vast majority of available methods have been developed for data stream classification and only a few of them attempted to solve regression problems, using various heuristic approaches. In this paper, we develop mathematically justified regression models working in a time-varying environment. More specifically, we study incremental versions of generalized regression neural networks, called IGRNNs, and we prove their tracking properties - weak (in probability) and strong (with probability one) convergence assuming various concept drift scenarios. First, we present the IGRNNs, based on the Parzen kernels, for modeling stationary systems under nonstationary noise. Next, we extend our approach to modeling time-varying systems under nonstationary noise. We present several types of concept drifts to be handled by our approach in such a way that weak and strong convergence holds under certain conditions. Finally, in the series of simulations, we compare our method with commonly used heuristic approaches, based on forgetting mechanism or sliding windows, to deal with concept drift. Finally, we apply our concept in a real life scenario solving the problem of currency exchange rates prediction.

Comparing machine learning and logistic regression methods for predicting hypertension using a combination of gene expression and next-generation sequencing data.

PubMed

Held, Elizabeth; Cape, Joshua; Tintle, Nathan

2016-01-01

Machine learning methods continue to show promise in the analysis of data from genetic association studies because of the high number of variables relative to the number of observations. However, few best practices exist for the application of these methods. We extend a recently proposed supervised machine learning approach for predicting disease risk by genotypes to be able to incorporate gene expression data and rare variants. We then apply 2 different versions of the approach (radial and linear support vector machines) to simulated data from Genetic Analysis Workshop 19 and compare performance to logistic regression. Method performance was not radically different across the 3 methods, although the linear support vector machine tended to show small gains in predictive ability relative to a radial support vector machine and logistic regression. Importantly, as the number of genes in the models was increased, even when those genes contained causal rare variants, model predictive ability showed a statistically significant decrease in performance for both the radial support vector machine and logistic regression. The linear support vector machine showed more robust performance to the inclusion of additional genes. Further work is needed to evaluate machine learning approaches on larger samples and to evaluate the relative improvement in model prediction from the incorporation of gene expression data.

The association between short interpregnancy interval and preterm birth in Louisiana: a comparison of methods.

PubMed

Howard, Elizabeth J; Harville, Emily; Kissinger, Patricia; Xiong, Xu

2013-07-01

There is growing interest in the application of propensity scores (PS) in epidemiologic studies, especially within the field of reproductive epidemiology. This retrospective cohort study assesses the impact of a short interpregnancy interval (IPI) on preterm birth and compares the results of the conventional logistic regression analysis with analyses utilizing a PS. The study included 96,378 singleton infants from Louisiana birth certificate data (1995-2007). Five regression models designed for methods comparison are presented. Ten percent (10.17 %) of all births were preterm; 26.83 % of births were from a short IPI. The PS-adjusted model produced a more conservative estimate of the exposure variable compared to the conventional logistic regression method (β-coefficient: 0.21 vs. 0.43), as well as a smaller standard error (0.024 vs. 0.028), odds ratio and 95 % confidence intervals [1.15 (1.09, 1.20) vs. 1.23 (1.17, 1.30)]. The inclusion of more covariate and interaction terms in the PS did not change the estimates of the exposure variable. This analysis indicates that PS-adjusted regression may be appropriate for validation of conventional methods in a large dataset with a fairly common outcome. PS's may be beneficial in producing more precise estimates, especially for models with many confounders and effect modifiers and where conventional adjustment with logistic regression is unsatisfactory. Short intervals between pregnancies are associated with preterm birth in this population, according to either technique. Birth spacing is an issue that women have some control over. Educational interventions, including birth control, should be applied during prenatal visits and following delivery.

Prediction of silicon oxynitride plasma etching using a generalized regression neural network

NASA Astrophysics Data System (ADS)

Kim, Byungwhan; Lee, Byung Teak

2005-08-01

A prediction model of silicon oxynitride (SiON) etching was constructed using a neural network. Model prediction performance was improved by means of genetic algorithm. The etching was conducted in a C2F6 inductively coupled plasma. A 24 full factorial experiment was employed to systematically characterize parameter effects on SiON etching. The process parameters include radio frequency source power, bias power, pressure, and C2F6 flow rate. To test the appropriateness of the trained model, additional 16 experiments were conducted. For comparison, four types of statistical regression models were built. Compared to the best regression model, the optimized neural network model demonstrated an improvement of about 52%. The optimized model was used to infer etch mechanisms as a function of parameters. The pressure effect was noticeably large only as relatively large ion bombardment was maintained in the process chamber. Ion-bombardment-activated polymer deposition played the most significant role in interpreting the complex effect of bias power or C2F6 flow rate. Moreover, [CF2] was expected to be the predominant precursor to polymer deposition.

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.

Modeling time-to-event (survival) data using classification tree analysis.

PubMed

Linden, Ariel; Yarnold, Paul R

2017-12-01

Time to the occurrence of an event is often studied in health research. Survival analysis differs from other designs in that follow-up times for individuals who do not experience the event by the end of the study (called censored) are accounted for in the analysis. Cox regression is the standard method for analysing censored data, but the assumptions required of these models are easily violated. In this paper, we introduce classification tree analysis (CTA) as a flexible alternative for modelling censored data. Classification tree analysis is a "decision-tree"-like classification model that provides parsimonious, transparent (ie, easy to visually display and interpret) decision rules that maximize predictive accuracy, derives exact P values via permutation tests, and evaluates model cross-generalizability. Using empirical data, we identify all statistically valid, reproducible, longitudinally consistent, and cross-generalizable CTA survival models and then compare their predictive accuracy to estimates derived via Cox regression and an unadjusted naïve model. Model performance is assessed using integrated Brier scores and a comparison between estimated survival curves. The Cox regression model best predicts average incidence of the outcome over time, whereas CTA survival models best predict either relatively high, or low, incidence of the outcome over time. Classification tree analysis survival models offer many advantages over Cox regression, such as explicit maximization of predictive accuracy, parsimony, statistical robustness, and transparency. Therefore, researchers interested in accurate prognoses and clear decision rules should consider developing models using the CTA-survival framework. © 2017 John Wiley & Sons, Ltd.

Locally-constrained Boundary Regression for Segmentation of Prostate and Rectum in the Planning CT Images

PubMed Central

Shao, Yeqin; Gao, Yaozong; Wang, Qian; Yang, Xin; Shen, Dinggang

2015-01-01

Automatic and accurate segmentation of the prostate and rectum in planning CT images is a challenging task due to low image contrast, unpredictable organ (relative) position, and uncertain existence of bowel gas across different patients. Recently, regression forest was adopted for organ deformable segmentation on 2D medical images by training one landmark detector for each point on the shape model. However, it seems impractical for regression forest to guide 3D deformable segmentation as a landmark detector, due to large number of vertices in the 3D shape model as well as the difficulty in building accurate 3D vertex correspondence for each landmark detector. In this paper, we propose a novel boundary detection method by exploiting the power of regression forest for prostate and rectum segmentation. The contributions of this paper are as follows: 1) we introduce regression forest as a local boundary regressor to vote the entire boundary of a target organ, which avoids training a large number of landmark detectors and building an accurate 3D vertex correspondence for each landmark detector; 2) an auto-context model is integrated with regression forest to improve the accuracy of the boundary regression; 3) we further combine a deformable segmentation method with the proposed local boundary regressor for the final organ segmentation by integrating organ shape priors. Our method is evaluated on a planning CT image dataset with 70 images from 70 different patients. The experimental results show that our proposed boundary regression method outperforms the conventional boundary classification method in guiding the deformable model for prostate and rectum segmentations. Compared with other state-of-the-art methods, our method also shows a competitive performance. PMID:26439938

Predicting Jakarta composite index using hybrid of fuzzy time series and support vector regression models

NASA Astrophysics Data System (ADS)

Febrian Umbara, Rian; Tarwidi, Dede; Budi Setiawan, Erwin

2018-03-01

The paper discusses the prediction of Jakarta Composite Index (JCI) in Indonesia Stock Exchange. The study is based on JCI historical data for 1286 days to predict the value of JCI one day ahead. This paper proposes predictions done in two stages., The first stage using Fuzzy Time Series (FTS) to predict values of ten technical indicators, and the second stage using Support Vector Regression (SVR) to predict the value of JCI one day ahead, resulting in a hybrid prediction model FTS-SVR. The performance of this combined prediction model is compared with the performance of the single stage prediction model using SVR only. Ten technical indicators are used as input for each model.

Modeling daily soil temperature over diverse climate conditions in Iran—a comparison of multiple linear regression and support vector regression techniques

NASA Astrophysics Data System (ADS)

Delbari, Masoomeh; Sharifazari, Salman; Mohammadi, Ehsan

2018-02-01

The knowledge of soil temperature at different depths is important for agricultural industry and for understanding climate change. The aim of this study is to evaluate the performance of a support vector regression (SVR)-based model in estimating daily soil temperature at 10, 30 and 100 cm depth at different climate conditions over Iran. The obtained results were compared to those obtained from a more classical multiple linear regression (MLR) model. The correlation sensitivity for the input combinations and periodicity effect were also investigated. Climatic data used as inputs to the models were minimum and maximum air temperature, solar radiation, relative humidity, dew point, and the atmospheric pressure (reduced to see level), collected from five synoptic stations Kerman, Ahvaz, Tabriz, Saghez, and Rasht located respectively in the hyper-arid, arid, semi-arid, Mediterranean, and hyper-humid climate conditions. According to the results, the performance of both MLR and SVR models was quite well at surface layer, i.e., 10-cm depth. However, SVR performed better than MLR in estimating soil temperature at deeper layers especially 100 cm depth. Moreover, both models performed better in humid climate condition than arid and hyper-arid areas. Further, adding a periodicity component into the modeling process considerably improved the models' performance especially in the case of SVR.

Cost-sensitive AdaBoost algorithm for ordinal regression based on extreme learning machine.

PubMed

Riccardi, Annalisa; Fernández-Navarro, Francisco; Carloni, Sante

2014-10-01

In this paper, the well known stagewise additive modeling using a multiclass exponential (SAMME) boosting algorithm is extended to address problems where there exists a natural order in the targets using a cost-sensitive approach. The proposed ensemble model uses an extreme learning machine (ELM) model as a base classifier (with the Gaussian kernel and the additional regularization parameter). The closed form of the derived weighted least squares problem is provided, and it is employed to estimate analytically the parameters connecting the hidden layer to the output layer at each iteration of the boosting algorithm. Compared to the state-of-the-art boosting algorithms, in particular those using ELM as base classifier, the suggested technique does not require the generation of a new training dataset at each iteration. The adoption of the weighted least squares formulation of the problem has been presented as an unbiased and alternative approach to the already existing ELM boosting techniques. Moreover, the addition of a cost model for weighting the patterns, according to the order of the targets, enables the classifier to tackle ordinal regression problems further. The proposed method has been validated by an experimental study by comparing it with already existing ensemble methods and ELM techniques for ordinal regression, showing competitive results.

Finding gene clusters for a replicated time course study

PubMed Central

2014-01-01

Background Finding genes that share similar expression patterns across samples is an important question that is frequently asked in high-throughput microarray studies. Traditional clustering algorithms such as K-means clustering and hierarchical clustering base gene clustering directly on the observed measurements and do not take into account the specific experimental design under which the microarray data were collected. A new model-based clustering method, the clustering of regression models method, takes into account the specific design of the microarray study and bases the clustering on how genes are related to sample covariates. It can find useful gene clusters for studies from complicated study designs such as replicated time course studies. Findings In this paper, we applied the clustering of regression models method to data from a time course study of yeast on two genotypes, wild type and YOX1 mutant, each with two technical replicates, and compared the clustering results with K-means clustering. We identified gene clusters that have similar expression patterns in wild type yeast, two of which were missed by K-means clustering. We further identified gene clusters whose expression patterns were changed in YOX1 mutant yeast compared to wild type yeast. Conclusions The clustering of regression models method can be a valuable tool for identifying genes that are coordinately transcribed by a common mechanism. PMID:24460656

A comparative study of generalized linear mixed modelling and artificial neural network approach for the joint modelling of survival and incidence of Dengue patients in Sri Lanka

NASA Astrophysics Data System (ADS)

Hapugoda, J. C.; Sooriyarachchi, M. R.

2017-09-01

Survival time of patients with a disease and the incidence of that particular disease (count) is frequently observed in medical studies with the data of a clustered nature. In many cases, though, the survival times and the count can be correlated in a way that, diseases that occur rarely could have shorter survival times or vice versa. Due to this fact, joint modelling of these two variables will provide interesting and certainly improved results than modelling these separately. Authors have previously proposed a methodology using Generalized Linear Mixed Models (GLMM) by joining the Discrete Time Hazard model with the Poisson Regression model to jointly model survival and count model. As Aritificial Neural Network (ANN) has become a most powerful computational tool to model complex non-linear systems, it was proposed to develop a new joint model of survival and count of Dengue patients of Sri Lanka by using that approach. Thus, the objective of this study is to develop a model using ANN approach and compare the results with the previously developed GLMM model. As the response variables are continuous in nature, Generalized Regression Neural Network (GRNN) approach was adopted to model the data. To compare the model fit, measures such as root mean square error (RMSE), absolute mean error (AME) and correlation coefficient (R) were used. The measures indicate the GRNN model fits the data better than the GLMM model.

Individualized Prediction of Heat Stress in Firefighters: A Data-Driven Approach Using Classification and Regression Trees.

PubMed

Mani, Ashutosh; Rao, Marepalli; James, Kelley; Bhattacharya, Amit

2015-01-01

The purpose of this study was to explore data-driven models, based on decision trees, to develop practical and easy to use predictive models for early identification of firefighters who are likely to cross the threshold of hyperthermia during live-fire training. Predictive models were created for three consecutive live-fire training scenarios. The final predicted outcome was a categorical variable: will a firefighter cross the upper threshold of hyperthermia - Yes/No. Two tiers of models were built, one with and one without taking into account the outcome (whether a firefighter crossed hyperthermia or not) from the previous training scenario. First tier of models included age, baseline heart rate and core body temperature, body mass index, and duration of training scenario as predictors. The second tier of models included the outcome of the previous scenario in the prediction space, in addition to all the predictors from the first tier of models. Classification and regression trees were used independently for prediction. The response variable for the regression tree was the quantitative variable: core body temperature at the end of each scenario. The predicted quantitative variable from regression trees was compared to the upper threshold of hyperthermia (38°C) to predict whether a firefighter would enter hyperthermia. The performance of classification and regression tree models was satisfactory for the second (success rate = 79%) and third (success rate = 89%) training scenarios but not for the first (success rate = 43%). Data-driven models based on decision trees can be a useful tool for predicting physiological response without modeling the underlying physiological systems. Early prediction of heat stress coupled with proactive interventions, such as pre-cooling, can help reduce heat stress in firefighters.

Filtering data from the collaborative initial glaucoma treatment study for improved identification of glaucoma progression.

PubMed

Schell, Greggory J; Lavieri, Mariel S; Stein, Joshua D; Musch, David C

2013-12-21

Open-angle glaucoma (OAG) is a prevalent, degenerate ocular disease which can lead to blindness without proper clinical management. The tests used to assess disease progression are susceptible to process and measurement noise. The aim of this study was to develop a methodology which accounts for the inherent noise in the data and improve significant disease progression identification. Longitudinal observations from the Collaborative Initial Glaucoma Treatment Study (CIGTS) were used to parameterize and validate a Kalman filter model and logistic regression function. The Kalman filter estimates the true value of biomarkers associated with OAG and forecasts future values of these variables. We develop two logistic regression models via generalized estimating equations (GEE) for calculating the probability of experiencing significant OAG progression: one model based on the raw measurements from CIGTS and another model based on the Kalman filter estimates of the CIGTS data. Receiver operating characteristic (ROC) curves and associated area under the ROC curve (AUC) estimates are calculated using cross-fold validation. The logistic regression model developed using Kalman filter estimates as data input achieves higher sensitivity and specificity than the model developed using raw measurements. The mean AUC for the Kalman filter-based model is 0.961 while the mean AUC for the raw measurements model is 0.889. Hence, using the probability function generated via Kalman filter estimates and GEE for logistic regression, we are able to more accurately classify patients and instances as experiencing significant OAG progression. A Kalman filter approach for estimating the true value of OAG biomarkers resulted in data input which improved the accuracy of a logistic regression classification model compared to a model using raw measurements as input. This methodology accounts for process and measurement noise to enable improved discrimination between progression and nonprogression in chronic diseases.

Sources of Biased Inference in Alcohol and Drug Services Research: An Instrumental Variable Approach

PubMed Central

Schmidt, Laura A.; Tam, Tammy W.; Larson, Mary Jo

2012-01-01

Objective: This study examined the potential for biased inference due to endogeneity when using standard approaches for modeling the utilization of alcohol and drug treatment. Method: Results from standard regression analysis were compared with those that controlled for endogeneity using instrumental variables estimation. Comparable models predicted the likelihood of receiving alcohol treatment based on the widely used Aday and Andersen medical care–seeking model. Data were from the National Epidemiologic Survey on Alcohol and Related Conditions and included a representative sample of adults in households and group quarters throughout the contiguous United States. Results: Findings suggested that standard approaches for modeling treatment utilization are prone to bias because of uncontrolled reverse causation and omitted variables. Compared with instrumental variables estimation, standard regression analyses produced downwardly biased estimates of the impact of alcohol problem severity on the likelihood of receiving care. Conclusions: Standard approaches for modeling service utilization are prone to underestimating the true effects of problem severity on service use. Biased inference could lead to inaccurate policy recommendations, for example, by suggesting that people with milder forms of substance use disorder are more likely to receive care than is actually the case. PMID:22152672

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.

Temperature-viscosity models reassessed.

PubMed

Peleg, Micha

2017-05-04

The temperature effect on viscosity of liquid and semi-liquid foods has been traditionally described by the Arrhenius equation, a few other mathematical models, and more recently by the WLF and VTF (or VFT) equations. The essence of the Arrhenius equation is that the viscosity is proportional to the absolute temperature's reciprocal and governed by a single parameter, namely, the energy of activation. However, if the absolute temperature in K in the Arrhenius equation is replaced by T + b where both T and the adjustable b are in °C, the result is a two-parameter model, which has superior fit to experimental viscosity-temperature data. This modified version of the Arrhenius equation is also mathematically equal to the WLF and VTF equations, which are known to be equal to each other. Thus, despite their dissimilar appearances all three equations are essentially the same model, and when used to fit experimental temperature-viscosity data render exactly the same very high regression coefficient. It is shown that three new hybrid two-parameter mathematical models, whose formulation bears little resemblance to any of the conventional models, can also have excellent fit with r 2 ∼ 1. This is demonstrated by comparing the various models' regression coefficients to published viscosity-temperature relationships of 40% sucrose solution, soybean oil, and 70°Bx pear juice concentrate at different temperature ranges. Also compared are reconstructed temperature-viscosity curves using parameters calculated directly from 2 or 3 data points and fitted curves obtained by nonlinear regression using a larger number of experimental viscosity measurements.

Comparison of linear and non-linear models for predicting energy expenditure from raw accelerometer data.

PubMed

Montoye, Alexander H K; Begum, Munni; Henning, Zachary; Pfeiffer, Karin A

2017-02-01

This study had three purposes, all related to evaluating energy expenditure (EE) prediction accuracy from body-worn accelerometers: (1) compare linear regression to linear mixed models, (2) compare linear models to artificial neural network models, and (3) compare accuracy of accelerometers placed on the hip, thigh, and wrists. Forty individuals performed 13 activities in a 90 min semi-structured, laboratory-based protocol. Participants wore accelerometers on the right hip, right thigh, and both wrists and a portable metabolic analyzer (EE criterion). Four EE prediction models were developed for each accelerometer: linear regression, linear mixed, and two ANN models. EE prediction accuracy was assessed using correlations, root mean square error (RMSE), and bias and was compared across models and accelerometers using repeated-measures analysis of variance. For all accelerometer placements, there were no significant differences for correlations or RMSE between linear regression and linear mixed models (correlations: r  =  0.71-0.88, RMSE: 1.11-1.61 METs; p  >  0.05). For the thigh-worn accelerometer, there were no differences in correlations or RMSE between linear and ANN models (ANN-correlations: r  =  0.89, RMSE: 1.07-1.08 METs. Linear models-correlations: r  =  0.88, RMSE: 1.10-1.11 METs; p  >  0.05). Conversely, one ANN had higher correlations and lower RMSE than both linear models for the hip (ANN-correlation: r  =  0.88, RMSE: 1.12 METs. Linear models-correlations: r  =  0.86, RMSE: 1.18-1.19 METs; p  <  0.05), and both ANNs had higher correlations and lower RMSE than both linear models for the wrist-worn accelerometers (ANN-correlations: r  =  0.82-0.84, RMSE: 1.26-1.32 METs. Linear models-correlations: r  =  0.71-0.73, RMSE: 1.55-1.61 METs; p  <  0.01). For studies using wrist-worn accelerometers, machine learning models offer a significant improvement in EE prediction accuracy over linear models. Conversely, linear models showed similar EE prediction accuracy to machine learning models for hip- and thigh-worn accelerometers and may be viable alternative modeling techniques for EE prediction for hip- or thigh-worn accelerometers.

Modeling thermal sensation in a Mediterranean climate—a comparison of linear and ordinal models

NASA Astrophysics Data System (ADS)

Pantavou, Katerina; Lykoudis, Spyridon

2014-08-01

A simple thermo-physiological model of outdoor thermal sensation adjusted with psychological factors is developed aiming to predict thermal sensation in Mediterranean climates. Microclimatic measurements simultaneously with interviews on personal and psychological conditions were carried out in a square, a street canyon and a coastal location of the greater urban area of Athens, Greece. Multiple linear and ordinal regression were applied in order to estimate thermal sensation making allowance for all the recorded parameters or specific, empirically selected, subsets producing so-called extensive and empirical models, respectively. Meteorological, thermo-physiological and overall models - considering psychological factors as well - were developed. Predictions were improved when personal and psychological factors were taken into account as compared to meteorological models. The model based on ordinal regression reproduced extreme values of thermal sensation vote more adequately than the linear regression one, while the empirical model produced satisfactory results in relation to the extensive model. The effects of adaptation and expectation on thermal sensation vote were introduced in the models by means of the exposure time, season and preference related to air temperature and irradiation. The assessment of thermal sensation could be a useful criterion in decision making regarding public health, outdoor spaces planning and tourism.

A Comparison of Alternative Approaches to the Analysis of Interrupted Time-Series.

ERIC Educational Resources Information Center

Harrop, John W.; Velicer, Wayne F.

1985-01-01

Computer generated data representative of 16 Auto Regressive Integrated Moving Averages (ARIMA) models were used to compare the results of interrupted time-series analysis using: (1) the known model identification, (2) an assumed (l,0,0) model, and (3) an assumed (3,0,0) model as an approximation to the General Transformation approach. (Author/BW)

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...

Classical Testing in Functional Linear Models.

PubMed

Kong, Dehan; Staicu, Ana-Maria; Maity, Arnab

2016-01-01

We extend four tests common in classical regression - Wald, score, likelihood ratio and F tests - to functional linear regression, for testing the null hypothesis, that there is no association between a scalar response and a functional covariate. Using functional principal component analysis, we re-express the functional linear model as a standard linear model, where the effect of the functional covariate can be approximated by a finite linear combination of the functional principal component scores. In this setting, we consider application of the four traditional tests. The proposed testing procedures are investigated theoretically for densely observed functional covariates when the number of principal components diverges. Using the theoretical distribution of the tests under the alternative hypothesis, we develop a procedure for sample size calculation in the context of functional linear regression. The four tests are further compared numerically for both densely and sparsely observed noisy functional data in simulation experiments and using two real data applications.

Classical Testing in Functional Linear Models

PubMed Central

Kong, Dehan; Staicu, Ana-Maria; Maity, Arnab

2016-01-01

We extend four tests common in classical regression - Wald, score, likelihood ratio and F tests - to functional linear regression, for testing the null hypothesis, that there is no association between a scalar response and a functional covariate. Using functional principal component analysis, we re-express the functional linear model as a standard linear model, where the effect of the functional covariate can be approximated by a finite linear combination of the functional principal component scores. In this setting, we consider application of the four traditional tests. The proposed testing procedures are investigated theoretically for densely observed functional covariates when the number of principal components diverges. Using the theoretical distribution of the tests under the alternative hypothesis, we develop a procedure for sample size calculation in the context of functional linear regression. The four tests are further compared numerically for both densely and sparsely observed noisy functional data in simulation experiments and using two real data applications. PMID:28955155

Modified retrieval algorithm for three types of precipitation distribution using x-band synthetic aperture radar

NASA Astrophysics Data System (ADS)

Xie, Yanan; Zhou, Mingliang; Pan, Dengke

2017-10-01

The forward-scattering model is introduced to describe the response of normalized radar cross section (NRCS) of precipitation with synthetic aperture radar (SAR). Since the distribution of near-surface rainfall is related to the rate of near-surface rainfall and horizontal distribution factor, a retrieval algorithm called modified regression empirical and model-oriented statistical (M-M) based on the volterra integration theory is proposed. Compared with the model-oriented statistical and volterra integration (MOSVI) algorithm, the biggest difference is that the M-M algorithm is based on the modified regression empirical algorithm rather than the linear regression formula to retrieve the value of near-surface rainfall rate. Half of the empirical parameters are reduced in the weighted integral work and a smaller average relative error is received while the rainfall rate is less than 100 mm/h. Therefore, the algorithm proposed in this paper can obtain high-precision rainfall information.

Machine Learning Algorithms Outperform Conventional Regression Models in Predicting Development of Hepatocellular Carcinoma

PubMed Central

Singal, Amit G.; Mukherjee, Ashin; Elmunzer, B. Joseph; Higgins, Peter DR; Lok, Anna S.; Zhu, Ji; Marrero, Jorge A; Waljee, Akbar K

2015-01-01

Background Predictive models for hepatocellular carcinoma (HCC) have been limited by modest accuracy and lack of validation. Machine learning algorithms offer a novel methodology, which may improve HCC risk prognostication among patients with cirrhosis. Our study's aim was to develop and compare predictive models for HCC development among cirrhotic patients, using conventional regression analysis and machine learning algorithms. Methods We enrolled 442 patients with Child A or B cirrhosis at the University of Michigan between January 2004 and September 2006 (UM cohort) and prospectively followed them until HCC development, liver transplantation, death, or study termination. Regression analysis and machine learning algorithms were used to construct predictive models for HCC development, which were tested on an independent validation cohort from the Hepatitis C Antiviral Long-term Treatment against Cirrhosis (HALT-C) Trial. Both models were also compared to the previously published HALT-C model. Discrimination was assessed using receiver operating characteristic curve analysis and diagnostic accuracy was assessed with net reclassification improvement and integrated discrimination improvement statistics. Results After a median follow-up of 3.5 years, 41 patients developed HCC. The UM regression model had a c-statistic of 0.61 (95%CI 0.56-0.67), whereas the machine learning algorithm had a c-statistic of 0.64 (95%CI 0.60–0.69) in the validation cohort. The machine learning algorithm had significantly better diagnostic accuracy as assessed by net reclassification improvement (p<0.001) and integrated discrimination improvement (p=0.04). The HALT-C model had a c-statistic of 0.60 (95%CI 0.50-0.70) in the validation cohort and was outperformed by the machine learning algorithm (p=0.047). Conclusion Machine learning algorithms improve the accuracy of risk stratifying patients with cirrhosis and can be used to accurately identify patients at high-risk for developing HCC. PMID:24169273

Monthly streamflow forecasting with auto-regressive integrated moving average

NASA Astrophysics Data System (ADS)

Nasir, Najah; Samsudin, Ruhaidah; Shabri, Ani

2017-09-01

Forecasting of streamflow is one of the many ways that can contribute to better decision making for water resource management. The auto-regressive integrated moving average (ARIMA) model was selected in this research for monthly streamflow forecasting with enhancement made by pre-processing the data using singular spectrum analysis (SSA). This study also proposed an extension of the SSA technique to include a step where clustering was performed on the eigenvector pairs before reconstruction of the time series. The monthly streamflow data of Sungai Muda at Jeniang, Sungai Muda at Jambatan Syed Omar and Sungai Ketil at Kuala Pegang was gathered from the Department of Irrigation and Drainage Malaysia. A ratio of 9:1 was used to divide the data into training and testing sets. The ARIMA, SSA-ARIMA and Clustered SSA-ARIMA models were all developed in R software. Results from the proposed model are then compared to a conventional auto-regressive integrated moving average model using the root-mean-square error and mean absolute error values. It was found that the proposed model can outperform the conventional model.

Adjustment of regional regression models of urban-runoff quality using data for Chattanooga, Knoxville, and Nashville, Tennessee

USGS Publications Warehouse

Hoos, Anne B.; Patel, Anant R.

1996-01-01

Model-adjustment procedures were applied to the combined data bases of storm-runoff quality for Chattanooga, Knoxville, and Nashville, Tennessee, to improve predictive accuracy for storm-runoff quality for urban watersheds in these three cities and throughout Middle and East Tennessee. Data for 45 storms at 15 different sites (five sites in each city) constitute the data base. Comparison of observed values of storm-runoff load and event-mean concentration to the predicted values from the regional regression models for 10 constituents shows prediction errors, as large as 806,000 percent. Model-adjustment procedures, which combine the regional model predictions with local data, are applied to improve predictive accuracy. Standard error of estimate after model adjustment ranges from 67 to 322 percent. Calibration results may be biased due to sampling error in the Tennessee data base. The relatively large values of standard error of estimate for some of the constituent models, although representing significant reduction (at least 50 percent) in prediction error compared to estimation with unadjusted regional models, may be unacceptable for some applications. The user may wish to collect additional local data for these constituents and repeat the analysis, or calibrate an independent local regression model.

Comparison of naïve Bayes and logistic regression for computer-aided diagnosis of breast masses using ultrasound imaging

NASA Astrophysics Data System (ADS)

Cary, Theodore W.; Cwanger, Alyssa; Venkatesh, Santosh S.; Conant, Emily F.; Sehgal, Chandra M.

2012-03-01

This study compares the performance of two proven but very different machine learners, Naïve Bayes and logistic regression, for differentiating malignant and benign breast masses using ultrasound imaging. Ultrasound images of 266 masses were analyzed quantitatively for shape, echogenicity, margin characteristics, and texture features. These features along with patient age, race, and mammographic BI-RADS category were used to train Naïve Bayes and logistic regression classifiers to diagnose lesions as malignant or benign. ROC analysis was performed using all of the features and using only a subset that maximized information gain. Performance was determined by the area under the ROC curve, Az, obtained from leave-one-out cross validation. Naïve Bayes showed significant variation (Az 0.733 +/- 0.035 to 0.840 +/- 0.029, P < 0.002) with the choice of features, but the performance of logistic regression was relatively unchanged under feature selection (Az 0.839 +/- 0.029 to 0.859 +/- 0.028, P = 0.605). Out of 34 features, a subset of 6 gave the highest information gain: brightness difference, margin sharpness, depth-to-width, mammographic BI-RADs, age, and race. The probabilities of malignancy determined by Naïve Bayes and logistic regression after feature selection showed significant correlation (R2= 0.87, P < 0.0001). The diagnostic performance of Naïve Bayes and logistic regression can be comparable, but logistic regression is more robust. Since probability of malignancy cannot be measured directly, high correlation between the probabilities derived from two basic but dissimilar models increases confidence in the predictive power of machine learning models for characterizing solid breast masses on ultrasound.

A classical regression framework for mediation analysis: fitting one model to estimate mediation effects.

PubMed

Saunders, Christina T; Blume, Jeffrey D

2017-10-26

Mediation analysis explores the degree to which an exposure's effect on an outcome is diverted through a mediating variable. We describe a classical regression framework for conducting mediation analyses in which estimates of causal mediation effects and their variance are obtained from the fit of a single regression model. The vector of changes in exposure pathway coefficients, which we named the essential mediation components (EMCs), is used to estimate standard causal mediation effects. Because these effects are often simple functions of the EMCs, an analytical expression for their model-based variance follows directly. Given this formula, it is instructive to revisit the performance of routinely used variance approximations (e.g., delta method and resampling methods). Requiring the fit of only one model reduces the computation time required for complex mediation analyses and permits the use of a rich suite of regression tools that are not easily implemented on a system of three equations, as would be required in the Baron-Kenny framework. Using data from the BRAIN-ICU study, we provide examples to illustrate the advantages of this framework and compare it with the existing approaches. © The Author 2017. Published by Oxford University Press.

An enhanced PM 2.5 air quality forecast model based on nonlinear regression and back-trajectory concentrations

NASA Astrophysics Data System (ADS)

Cobourn, W. Geoffrey

2010-08-01

An enhanced PM 2.5 air quality forecast model based on nonlinear regression (NLR) and back-trajectory concentrations has been developed for use in the Louisville, Kentucky metropolitan area. The PM 2.5 air quality forecast model is designed for use in the warm season, from May through September, when PM 2.5 air quality is more likely to be critical for human health. The enhanced PM 2.5 model consists of a basic NLR model, developed for use with an automated air quality forecast system, and an additional parameter based on upwind PM 2.5 concentration, called PM24. The PM24 parameter is designed to be determined manually, by synthesizing backward air trajectory and regional air quality information to compute 24-h back-trajectory concentrations. The PM24 parameter may be used by air quality forecasters to adjust the forecast provided by the automated forecast system. In this study of the 2007 and 2008 forecast seasons, the enhanced model performed well using forecasted meteorological data and PM24 as input. The enhanced PM 2.5 model was compared with three alternative models, including the basic NLR model, the basic NLR model with a persistence parameter added, and the NLR model with persistence and PM24. The two models that included PM24 were of comparable accuracy. The two models incorporating back-trajectory concentrations had lower mean absolute errors and higher rates of detecting unhealthy PM2.5 concentrations compared to the other models.

Aggregating the response in time series regression models, applied to weather-related cardiovascular mortality.

PubMed

Masselot, Pierre; Chebana, Fateh; Bélanger, Diane; St-Hilaire, André; Abdous, Belkacem; Gosselin, Pierre; Ouarda, Taha B M J

2018-07-01

In environmental epidemiology studies, health response data (e.g. hospitalization or mortality) are often noisy because of hospital organization and other social factors. The noise in the data can hide the true signal related to the exposure. The signal can be unveiled by performing a temporal aggregation on health data and then using it as the response in regression analysis. From aggregated series, a general methodology is introduced to account for the particularities of an aggregated response in a regression setting. This methodology can be used with usually applied regression models in weather-related health studies, such as generalized additive models (GAM) and distributed lag nonlinear models (DLNM). In particular, the residuals are modelled using an autoregressive-moving average (ARMA) model to account for the temporal dependence. The proposed methodology is illustrated by modelling the influence of temperature on cardiovascular mortality in Canada. A comparison with classical DLNMs is provided and several aggregation methods are compared. Results show that there is an increase in the fit quality when the response is aggregated, and that the estimated relationship focuses more on the outcome over several days than the classical DLNM. More precisely, among various investigated aggregation schemes, it was found that an aggregation with an asymmetric Epanechnikov kernel is more suited for studying the temperature-mortality relationship. Copyright © 2018. Published by Elsevier B.V.

Optimizing separate phase light hydrocarbon recovery from contaminated unconfined aquifers

NASA Astrophysics Data System (ADS)

Cooper, Grant S.; Peralta, Richard C.; Kaluarachchi, Jagath J.

A modeling approach is presented that optimizes separate phase recovery of light non-aqueous phase liquids (LNAPL) for a single dual-extraction well in a homogeneous, isotropic unconfined aquifer. A simulation/regression/optimization (S/R/O) model is developed to predict, analyze, and optimize the oil recovery process. The approach combines detailed simulation, nonlinear regression, and optimization. The S/R/O model utilizes nonlinear regression equations describing system response to time-varying water pumping and oil skimming. Regression equations are developed for residual oil volume and free oil volume. The S/R/O model determines optimized time-varying (stepwise) pumping rates which minimize residual oil volume and maximize free oil recovery while causing free oil volume to decrease a specified amount. This S/R/O modeling approach implicitly immobilizes the free product plume by reversing the water table gradient while achieving containment. Application to a simple representative problem illustrates the S/R/O model utility for problem analysis and remediation design. When compared with the best steady pumping strategies, the optimal stepwise pumping strategy improves free oil recovery by 11.5% and reduces the amount of residual oil left in the system due to pumping by 15%. The S/R/O model approach offers promise for enhancing the design of free phase LNAPL recovery systems and to help in making cost-effective operation and management decisions for hydrogeologists, engineers, and regulators.

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.

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

Methods for estimating the magnitude and frequency of peak streamflows at ungaged sites in and near the Oklahoma Panhandle

USGS Publications Warehouse

Smith, S. Jerrod; Lewis, Jason M.; Graves, Grant M.

2015-09-28

Generalized-least-squares multiple-linear regression analysis was used to formulate regression relations between peak-streamflow frequency statistics and basin characteristics. Contributing drainage area was the only basin characteristic determined to be statistically significant for all percentage of annual exceedance probabilities and was the only basin characteristic used in regional regression equations for estimating peak-streamflow frequency statistics on unregulated streams in and near the Oklahoma Panhandle. The regression model pseudo-coefficient of determination, converted to percent, for the Oklahoma Panhandle regional regression equations ranged from about 38 to 63 percent. The standard errors of prediction and the standard model errors for the Oklahoma Panhandle regional regression equations ranged from about 84 to 148 percent and from about 76 to 138 percent, respectively. These errors were comparable to those reported for regional peak-streamflow frequency regression equations for the High Plains areas of Texas and Colorado. The root mean square errors for the Oklahoma Panhandle regional regression equations (ranging from 3,170 to 92,000 cubic feet per second) were less than the root mean square errors for the Oklahoma statewide regression equations (ranging from 18,900 to 412,000 cubic feet per second); therefore, the Oklahoma Panhandle regional regression equations produce more accurate peak-streamflow statistic estimates for the irrigated period of record in the Oklahoma Panhandle than do the Oklahoma statewide regression equations. The regression equations developed in this report are applicable to streams that are not substantially affected by regulation, impoundment, or surface-water withdrawals. These regression equations are intended for use for stream sites with contributing drainage areas less than or equal to about 2,060 square miles, the maximum value for the independent variable used in the regression analysis.

Multiplication factor versus regression analysis in stature estimation from hand and foot dimensions.

PubMed

Krishan, Kewal; Kanchan, Tanuj; Sharma, Abhilasha

2012-05-01

Estimation of stature is an important parameter in identification of human remains in forensic examinations. The present study is aimed to compare the reliability and accuracy of stature estimation and to demonstrate the variability in estimated stature and actual stature using multiplication factor and regression analysis methods. The study is based on a sample of 246 subjects (123 males and 123 females) from North India aged between 17 and 20 years. Four anthropometric measurements; hand length, hand breadth, foot length and foot breadth taken on the left side in each subject were included in the study. Stature was measured using standard anthropometric techniques. Multiplication factors were calculated and linear regression models were derived for estimation of stature from hand and foot dimensions. Derived multiplication factors and regression formula were applied to the hand and foot measurements in the study sample. The estimated stature from the multiplication factors and regression analysis was compared with the actual stature to find the error in estimated stature. The results indicate that the range of error in estimation of stature from regression analysis method is less than that of multiplication factor method thus, confirming that the regression analysis method is better than multiplication factor analysis in stature estimation. Copyright © 2012 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved.

Simulation of groundwater level variations using wavelet combined with neural network, linear regression and support vector machine

NASA Astrophysics Data System (ADS)

Ebrahimi, Hadi; Rajaee, Taher

2017-01-01

Simulation of groundwater level (GWL) fluctuations is an important task in management of groundwater resources. In this study, the effect of wavelet analysis on the training of the artificial neural network (ANN), multi linear regression (MLR) and support vector regression (SVR) approaches was investigated, and the ANN, MLR and SVR along with the wavelet-ANN (WNN), wavelet-MLR (WLR) and wavelet-SVR (WSVR) models were compared in simulating one-month-ahead of GWL. The only variable used to develop the models was the monthly GWL data recorded over a period of 11 years from two wells in the Qom plain, Iran. The results showed that decomposing GWL time series into several sub-time series, extremely improved the training of the models. For both wells 1 and 2, the Meyer and Db5 wavelets produced better results compared to the other wavelets; which indicated wavelet types had similar behavior in similar case studies. The optimal number of delays was 6 months, which seems to be due to natural phenomena. The best WNN model, using Meyer mother wavelet with two decomposition levels, simulated one-month-ahead with RMSE values being equal to 0.069 m and 0.154 m for wells 1 and 2, respectively. The RMSE values for the WLR model were 0.058 m and 0.111 m, and for WSVR model were 0.136 m and 0.060 m for wells 1 and 2, respectively.

Comparison of methods for the prediction of human clearance from hepatocyte intrinsic clearance for a set of reference compounds and an external evaluation set.

PubMed

Yamagata, Tetsuo; Zanelli, Ugo; Gallemann, Dieter; Perrin, Dominique; Dolgos, Hugues; Petersson, Carl

2017-09-01

1. We compared direct scaling, regression model equation and the so-called "Poulin et al." methods to scale clearance (CL) from in vitro intrinsic clearance (CL int ) measured in human hepatocytes using two sets of compounds. One reference set comprised of 20 compounds with known elimination pathways and one external evaluation set based on 17 compounds development in Merck (MS). 2. A 90% prospective confidence interval was calculated using the reference set. This interval was found relevant for the regression equation method. The three outliers identified were justified on the basis of their elimination mechanism. 3. The direct scaling method showed a systematic underestimation of clearance in both the reference and evaluation sets. The "Poulin et al." and the regression equation methods showed no obvious bias in either the reference or evaluation sets. 4. The regression model equation was slightly superior to the "Poulin et al." method in the reference set and showed a better absolute average fold error (AAFE) of value 1.3 compared to 1.6. A larger difference was observed in the evaluation set were the regression method and "Poulin et al." resulted in an AAFE of 1.7 and 2.6, respectively (removing the three compounds with known issues mentioned above). A similar pattern was observed for the correlation coefficient. Based on these data we suggest the regression equation method combined with a prospective confidence interval as the first choice for the extrapolation of human in vivo hepatic metabolic clearance from in vitro systems.

Using Multivariate Adaptive Regression Spline and Artificial Neural Network to Simulate Urbanization in Mumbai, India

NASA Astrophysics Data System (ADS)

Ahmadlou, M.; Delavar, M. R.; Tayyebi, A.; Shafizadeh-Moghadam, H.

2015-12-01

Land use change (LUC) models used for modelling urban growth are different in structure and performance. Local models divide the data into separate subsets and fit distinct models on each of the subsets. Non-parametric models are data driven and usually do not have a fixed model structure or model structure is unknown before the modelling process. On the other hand, global models perform modelling using all the available data. In addition, parametric models have a fixed structure before the modelling process and they are model driven. Since few studies have compared local non-parametric models with global parametric models, this study compares a local non-parametric model called multivariate adaptive regression spline (MARS), and a global parametric model called artificial neural network (ANN) to simulate urbanization in Mumbai, India. Both models determine the relationship between a dependent variable and multiple independent variables. We used receiver operating characteristic (ROC) to compare the power of the both models for simulating urbanization. Landsat images of 1991 (TM) and 2010 (ETM+) were used for modelling the urbanization process. The drivers considered for urbanization in this area were distance to urban areas, urban density, distance to roads, distance to water, distance to forest, distance to railway, distance to central business district, number of agricultural cells in a 7 by 7 neighbourhoods, and slope in 1991. The results showed that the area under the ROC curve for MARS and ANN was 94.77% and 95.36%, respectively. Thus, ANN performed slightly better than MARS to simulate urban areas in Mumbai, India.

Improving Consensus Scoring of Crowdsourced Data Using the Rasch Model: Development and Refinement of a Diagnostic Instrument.

PubMed

Brady, Christopher John; Mudie, Lucy Iluka; Wang, Xueyang; Guallar, Eliseo; Friedman, David Steven

2017-06-20

Diabetic retinopathy (DR) is a leading cause of vision loss in working age individuals worldwide. While screening is effective and cost effective, it remains underutilized, and novel methods are needed to increase detection of DR. This clinical validation study compared diagnostic gradings of retinal fundus photographs provided by volunteers on the Amazon Mechanical Turk (AMT) crowdsourcing marketplace with expert-provided gold-standard grading and explored whether determination of the consensus of crowdsourced classifications could be improved beyond a simple majority vote (MV) using regression methods. The aim of our study was to determine whether regression methods could be used to improve the consensus grading of data collected by crowdsourcing. A total of 1200 retinal images of individuals with diabetes mellitus from the Messidor public dataset were posted to AMT. Eligible crowdsourcing workers had at least 500 previously approved tasks with an approval rating of 99% across their prior submitted work. A total of 10 workers were recruited to classify each image as normal or abnormal. If half or more workers judged the image to be abnormal, the MV consensus grade was recorded as abnormal. Rasch analysis was then used to calculate worker ability scores in a random 50% training set, which were then used as weights in a regression model in the remaining 50% test set to determine if a more accurate consensus could be devised. Outcomes of interest were the percent correctly classified images, sensitivity, specificity, and area under the receiver operating characteristic (AUROC) for the consensus grade as compared with the expert grading provided with the dataset. Using MV grading, the consensus was correct in 75.5% of images (906/1200), with 75.5% sensitivity, 75.5% specificity, and an AUROC of 0.75 (95% CI 0.73-0.78). A logistic regression model using Rasch-weighted individual scores generated an AUROC of 0.91 (95% CI 0.88-0.93) compared with 0.89 (95% CI 0.86-92) for a model using unweighted scores (chi-square P value<.001). Setting a diagnostic cut-point to optimize sensitivity at 90%, 77.5% (465/600) were graded correctly, with 90.3% sensitivity, 68.5% specificity, and an AUROC of 0.79 (95% CI 0.76-0.83). Crowdsourced interpretations of retinal images provide rapid and accurate results as compared with a gold-standard grading. Creating a logistic regression model using Rasch analysis to weight crowdsourced classifications by worker ability improves accuracy of aggregated grades as compared with simple majority vote. ©Christopher John Brady, Lucy Iluka Mudie, Xueyang Wang, Eliseo Guallar, David Steven Friedman. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 20.06.2017.

Improving Global Models of Remotely Sensed Ocean Chlorophyll Content Using Partial Least Squares and Geographically Weighted Regression

NASA Astrophysics Data System (ADS)

Gholizadeh, H.; Robeson, S. M.

2015-12-01

Empirical models have been widely used to estimate global chlorophyll content from remotely sensed data. Here, we focus on the standard NASA empirical models that use blue-green band ratios. These band ratio ocean color (OC) algorithms are in the form of fourth-order polynomials and the parameters of these polynomials (i.e. coefficients) are estimated from the NASA bio-Optical Marine Algorithm Data set (NOMAD). Most of the points in this data set have been sampled from tropical and temperate regions. However, polynomial coefficients obtained from this data set are used to estimate chlorophyll content in all ocean regions with different properties such as sea-surface temperature, salinity, and downwelling/upwelling patterns. Further, the polynomial terms in these models are highly correlated. In sum, the limitations of these empirical models are as follows: 1) the independent variables within the empirical models, in their current form, are correlated (multicollinear), and 2) current algorithms are global approaches and are based on the spatial stationarity assumption, so they are independent of location. Multicollinearity problem is resolved by using partial least squares (PLS). PLS, which transforms the data into a set of independent components, can be considered as a combined form of principal component regression (PCR) and multiple regression. Geographically weighted regression (GWR) is also used to investigate the validity of spatial stationarity assumption. GWR solves a regression model over each sample point by using the observations within its neighbourhood. PLS results show that the empirical method underestimates chlorophyll content in high latitudes, including the Southern Ocean region, when compared to PLS (see Figure 1). Cluster analysis of GWR coefficients also shows that the spatial stationarity assumption in empirical models is not likely a valid assumption.

Data mining: Potential applications in research on nutrition and health.

PubMed

Batterham, Marijka; Neale, Elizabeth; Martin, Allison; Tapsell, Linda

2017-02-01

Data mining enables further insights from nutrition-related research, but caution is required. The aim of this analysis was to demonstrate and compare the utility of data mining methods in classifying a categorical outcome derived from a nutrition-related intervention. Baseline data (23 variables, 8 categorical) on participants (n = 295) in an intervention trial were used to classify participants in terms of meeting the criteria of achieving 10 000 steps per day. Results from classification and regression trees (CARTs), random forests, adaptive boosting, logistic regression, support vector machines and neural networks were compared using area under the curve (AUC) and error assessments. The CART produced the best model when considering the AUC (0.703), overall error (18%) and within class error (28%). Logistic regression also performed reasonably well compared to the other models (AUC 0.675, overall error 23%, within class error 36%). All the methods gave different rankings of variables' importance. CART found that body fat, quality of life using the SF-12 Physical Component Summary (PCS) and the cholesterol: HDL ratio were the most important predictors of meeting the 10 000 steps criteria, while logistic regression showed the SF-12PCS, glucose levels and level of education to be the most significant predictors (P ≤ 0.01). Differing outcomes suggest caution is required with a single data mining method, particularly in a dataset with nonlinear relationships and outliers and when exploring relationships that were not the primary outcomes of the research. © 2017 Dietitians Association of Australia.

Post-processing through linear regression

NASA Astrophysics Data System (ADS)

van Schaeybroeck, B.; Vannitsem, S.

2011-03-01

Various post-processing techniques are compared for both deterministic and ensemble forecasts, all based on linear regression between forecast data and observations. In order to evaluate the quality of the regression methods, three criteria are proposed, related to the effective correction of forecast error, the optimal variability of the corrected forecast and multicollinearity. The regression schemes under consideration include the ordinary least-square (OLS) method, a new time-dependent Tikhonov regularization (TDTR) method, the total least-square method, a new geometric-mean regression (GM), a recently introduced error-in-variables (EVMOS) method and, finally, a "best member" OLS method. The advantages and drawbacks of each method are clarified. These techniques are applied in the context of the 63 Lorenz system, whose model version is affected by both initial condition and model errors. For short forecast lead times, the number and choice of predictors plays an important role. Contrarily to the other techniques, GM degrades when the number of predictors increases. At intermediate lead times, linear regression is unable to provide corrections to the forecast and can sometimes degrade the performance (GM and the best member OLS with noise). At long lead times the regression schemes (EVMOS, TDTR) which yield the correct variability and the largest correlation between ensemble error and spread, should be preferred.

Area-to-point regression kriging for pan-sharpening

NASA Astrophysics Data System (ADS)

Wang, Qunming; Shi, Wenzhong; Atkinson, Peter M.

2016-04-01

Pan-sharpening is a technique to combine the fine spatial resolution panchromatic (PAN) band with the coarse spatial resolution multispectral bands of the same satellite to create a fine spatial resolution multispectral image. In this paper, area-to-point regression kriging (ATPRK) is proposed for pan-sharpening. ATPRK considers the PAN band as the covariate. Moreover, ATPRK is extended with a local approach, called adaptive ATPRK (AATPRK), which fits a regression model using a local, non-stationary scheme such that the regression coefficients change across the image. The two geostatistical approaches, ATPRK and AATPRK, were compared to the 13 state-of-the-art pan-sharpening approaches summarized in Vivone et al. (2015) in experiments on three separate datasets. ATPRK and AATPRK produced more accurate pan-sharpened images than the 13 benchmark algorithms in all three experiments. Unlike the benchmark algorithms, the two geostatistical solutions precisely preserved the spectral properties of the original coarse data. Furthermore, ATPRK can be enhanced by a local scheme in AATRPK, in cases where the residuals from a global regression model are such that their spatial character varies locally.

Does Nonlinear Modeling Play a Role in Plasmid Bioprocess Monitoring Using Fourier Transform Infrared Spectra?

PubMed

Lopes, Marta B; Calado, Cecília R C; Figueiredo, Mário A T; Bioucas-Dias, José M

2017-06-01

The monitoring of biopharmaceutical products using Fourier transform infrared (FT-IR) spectroscopy relies on calibration techniques involving the acquisition of spectra of bioprocess samples along the process. The most commonly used method for that purpose is partial least squares (PLS) regression, under the assumption that a linear model is valid. Despite being successful in the presence of small nonlinearities, linear methods may fail in the presence of strong nonlinearities. This paper studies the potential usefulness of nonlinear regression methods for predicting, from in situ near-infrared (NIR) and mid-infrared (MIR) spectra acquired in high-throughput mode, biomass and plasmid concentrations in Escherichia coli DH5-α cultures producing the plasmid model pVAX-LacZ. The linear methods PLS and ridge regression (RR) are compared with their kernel (nonlinear) versions, kPLS and kRR, as well as with the (also nonlinear) relevance vector machine (RVM) and Gaussian process regression (GPR). For the systems studied, RR provided better predictive performances compared to the remaining methods. Moreover, the results point to further investigation based on larger data sets whenever differences in predictive accuracy between a linear method and its kernelized version could not be found. The use of nonlinear methods, however, shall be judged regarding the additional computational cost required to tune their additional parameters, especially when the less computationally demanding linear methods herein studied are able to successfully monitor the variables under study.

Mechanisms behind the estimation of photosynthesis traits from leaf reflectance observations

NASA Astrophysics Data System (ADS)

Dechant, Benjamin; Cuntz, Matthias; Doktor, Daniel; Vohland, Michael

2016-04-01

Many studies have investigated the reflectance-based estimation of leaf chlorophyll, water and dry matter contents of plants. Only few studies focused on photosynthesis traits, however. The maximum potential uptake of carbon dioxide under given environmental conditions is determined mainly by RuBisCO activity, limiting carboxylation, or the speed of photosynthetic electron transport. These two main limitations are represented by the maximum carboxylation capacity, V cmax,25, and the maximum electron transport rate, Jmax,25. These traits were estimated from leaf reflectance before but the mechanisms underlying the estimation remain rather speculative. The aim of this study was therefore to reveal the mechanisms behind reflectance-based estimation of V cmax,25 and Jmax,25. Leaf reflectance, photosynthetic response curves as well as nitrogen content per area, Narea, and leaf mass per area, LMA, were measured on 37 deciduous tree species. V cmax,25 and Jmax,25 were determined from the response curves. Partial Least Squares (PLS) regression models for the two photosynthesis traits V cmax,25 and Jmax,25 as well as Narea and LMA were studied using a cross-validation approach. Analyses of linear regression models based on Narea and other leaf traits estimated via PROSPECT inversion, PLS regression coefficients and model residuals were conducted in order to reveal the mechanisms behind the reflectance-based estimation. We found that V cmax,25 and Jmax,25 can be estimated from leaf reflectance with good to moderate accuracy for a large number of species and different light conditions. The dominant mechanism behind the estimations was the strong relationship between photosynthesis traits and leaf nitrogen content. This was concluded from very strong relationships between PLS regression coefficients, the model residuals as well as the prediction performance of Narea- based linear regression models compared to PLS regression models. While the PLS regression model for V cmax,25 was fully based on the correlation to Narea, the PLS regression model for Jmax,25 was not entirely based on it. Analyses of the contributions of different parts of the reflectance spectrum revealed that the information contributing to the Jmax,25 PLS regression model in addition to the main source of information, Narea, was mainly located in the visible part of the spectrum (500-900 nm). Estimated chlorophyll content could be excluded as potential source of this extra information. The PLS regression coefficients of the Jmax,25 model indicated possible contributions from chlorophyll fluorescence and cytochrome f content. In summary, we found that the main mechanism behind the estimation of V cmax,25 and Jmax,25 from leaf reflectance observations is the correlation to Narea but that there is additional information related to Jmax,25 mainly in the visible part of the spectrum.

Uncovering state-dependent relationships in shallow lakes using Bayesian latent variable regression.

PubMed

Vitense, Kelsey; Hanson, Mark A; Herwig, Brian R; Zimmer, Kyle D; Fieberg, John

2018-03-01

Ecosystems sometimes undergo dramatic shifts between contrasting regimes. Shallow lakes, for instance, can transition between two alternative stable states: a clear state dominated by submerged aquatic vegetation and a turbid state dominated by phytoplankton. Theoretical models suggest that critical nutrient thresholds differentiate three lake types: highly resilient clear lakes, lakes that may switch between clear and turbid states following perturbations, and highly resilient turbid lakes. For effective and efficient management of shallow lakes and other systems, managers need tools to identify critical thresholds and state-dependent relationships between driving variables and key system features. Using shallow lakes as a model system for which alternative stable states have been demonstrated, we developed an integrated framework using Bayesian latent variable regression (BLR) to classify lake states, identify critical total phosphorus (TP) thresholds, and estimate steady state relationships between TP and chlorophyll a (chl a) using cross-sectional data. We evaluated the method using data simulated from a stochastic differential equation model and compared its performance to k-means clustering with regression (KMR). We also applied the framework to data comprising 130 shallow lakes. For simulated data sets, BLR had high state classification rates (median/mean accuracy >97%) and accurately estimated TP thresholds and state-dependent TP-chl a relationships. Classification and estimation improved with increasing sample size and decreasing noise levels. Compared to KMR, BLR had higher classification rates and better approximated the TP-chl a steady state relationships and TP thresholds. We fit the BLR model to three different years of empirical shallow lake data, and managers can use the estimated bifurcation diagrams to prioritize lakes for management according to their proximity to thresholds and chance of successful rehabilitation. Our model improves upon previous methods for shallow lakes because it allows classification and regression to occur simultaneously and inform one another, directly estimates TP thresholds and the uncertainty associated with thresholds and state classifications, and enables meaningful constraints to be built into models. The BLR framework is broadly applicable to other ecosystems known to exhibit alternative stable states in which regression can be used to establish relationships between driving variables and state variables. © 2017 by the Ecological Society of America.

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.

Tutorial on Biostatistics: Linear Regression Analysis of Continuous Correlated Eye Data.

PubMed

Ying, Gui-Shuang; Maguire, Maureen G; Glynn, Robert; Rosner, Bernard

2017-04-01

To describe and demonstrate appropriate linear regression methods for analyzing correlated continuous eye data. We describe several approaches to regression analysis involving both eyes, including mixed effects and marginal models under various covariance structures to account for inter-eye correlation. We demonstrate, with SAS statistical software, applications in a study comparing baseline refractive error between one eye with choroidal neovascularization (CNV) and the unaffected fellow eye, and in a study determining factors associated with visual field in the elderly. When refractive error from both eyes were analyzed with standard linear regression without accounting for inter-eye correlation (adjusting for demographic and ocular covariates), the difference between eyes with CNV and fellow eyes was 0.15 diopters (D; 95% confidence interval, CI -0.03 to 0.32D, p = 0.10). Using a mixed effects model or a marginal model, the estimated difference was the same but with narrower 95% CI (0.01 to 0.28D, p = 0.03). Standard regression for visual field data from both eyes provided biased estimates of standard error (generally underestimated) and smaller p-values, while analysis of the worse eye provided larger p-values than mixed effects models and marginal models. In research involving both eyes, ignoring inter-eye correlation can lead to invalid inferences. Analysis using only right or left eyes is valid, but decreases power. Worse-eye analysis can provide less power and biased estimates of effect. Mixed effects or marginal models using the eye as the unit of analysis should be used to appropriately account for inter-eye correlation and maximize power and precision.

Multivariate decoding of brain images using ordinal regression.

PubMed

Doyle, O M; Ashburner, J; Zelaya, F O; Williams, S C R; Mehta, M A; Marquand, A F

2013-11-01

Neuroimaging data are increasingly being used to predict potential outcomes or groupings, such as clinical severity, drug dose response, and transitional illness states. In these examples, the variable (target) we want to predict is ordinal in nature. Conventional classification schemes assume that the targets are nominal and hence ignore their ranked nature, whereas parametric and/or non-parametric regression models enforce a metric notion of distance between classes. Here, we propose a novel, alternative multivariate approach that overcomes these limitations - whole brain probabilistic ordinal regression using a Gaussian process framework. We applied this technique to two data sets of pharmacological neuroimaging data from healthy volunteers. The first study was designed to investigate the effect of ketamine on brain activity and its subsequent modulation with two compounds - lamotrigine and risperidone. The second study investigates the effect of scopolamine on cerebral blood flow and its modulation using donepezil. We compared ordinal regression to multi-class classification schemes and metric regression. Considering the modulation of ketamine with lamotrigine, we found that ordinal regression significantly outperformed multi-class classification and metric regression in terms of accuracy and mean absolute error. However, for risperidone ordinal regression significantly outperformed metric regression but performed similarly to multi-class classification both in terms of accuracy and mean absolute error. For the scopolamine data set, ordinal regression was found to outperform both multi-class and metric regression techniques considering the regional cerebral blood flow in the anterior cingulate cortex. Ordinal regression was thus the only method that performed well in all cases. Our results indicate the potential of an ordinal regression approach for neuroimaging data while providing a fully probabilistic framework with elegant approaches for model selection. Copyright © 2013. Published by Elsevier Inc.

Exposure reconstruction for the TCDD-exposed NIOSH cohort using a concentration- and age-dependent model of elimination.

PubMed

Aylward, Lesa L; Brunet, Robert C; Starr, Thomas B; Carrier, Gaétan; Delzell, Elizabeth; Cheng, Hong; Beall, Colleen

2005-08-01

Recent studies demonstrating a concentration dependence of elimination of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) suggest that previous estimates of exposure for occupationally exposed cohorts may have underestimated actual exposure, resulting in a potential overestimate of the carcinogenic potency of TCDD in humans based on the mortality data for these cohorts. Using a database on U.S. chemical manufacturing workers potentially exposed to TCDD compiled by the National Institute for Occupational Safety and Health (NIOSH), we evaluated the impact of using a concentration- and age-dependent elimination model (CADM) (Aylward et al., 2005) on estimates of serum lipid area under the curve (AUC) for the NIOSH cohort. These data were used previously by Steenland et al. (2001) in combination with a first-order elimination model with an 8.7-year half-life to estimate cumulative serum lipid concentration (equivalent to AUC) for these workers for use in cancer dose-response assessment. Serum lipid TCDD measurements taken in 1988 for a subset of the cohort were combined with the NIOSH job exposure matrix and work histories to estimate dose rates per unit of exposure score. We evaluated the effect of choices in regression model (regression on untransformed vs. ln-transformed data and inclusion of a nonzero regression intercept) as well as the impact of choices of elimination models and parameters on estimated AUCs for the cohort. Central estimates for dose rate parameters derived from the serum-sampled subcohort were applied with the elimination models to time-specific exposure scores for the entire cohort to generate AUC estimates for all cohort members. Use of the CADM resulted in improved model fits to the serum sampling data compared to the first-order models. Dose rates varied by a factor of 50 among different combinations of elimination model, parameter sets, and regression models. Use of a CADM results in increases of up to five-fold in AUC estimates for the more highly exposed members of the cohort compared to estimates obtained using the first-order model with 8.7-year half-life. This degree of variation in the AUC estimates for this cohort would affect substantially the cancer potency estimates derived from the mortality data from this cohort. Such variability and uncertainty in the reconstructed serum lipid AUC estimates for this cohort, depending on elimination model, parameter set, and regression model, have not been described previously and are critical components in evaluating the dose-response data from the occupationally exposed populations.

Integration of logistic regression, Markov chain and cellular automata models to simulate urban expansion

NASA Astrophysics Data System (ADS)

Jokar Arsanjani, Jamal; Helbich, Marco; Kainz, Wolfgang; Darvishi Boloorani, Ali

2013-04-01

This research analyses the suburban expansion in the metropolitan area of Tehran, Iran. A hybrid model consisting of logistic regression model, Markov chain (MC), and cellular automata (CA) was designed to improve the performance of the standard logistic regression model. Environmental and socio-economic variables dealing with urban sprawl were operationalised to create a probability surface of spatiotemporal states of built-up land use for the years 2006, 2016, and 2026. For validation, the model was evaluated by means of relative operating characteristic values for different sets of variables. The approach was calibrated for 2006 by cross comparing of actual and simulated land use maps. The achieved outcomes represent a match of 89% between simulated and actual maps of 2006, which was satisfactory to approve the calibration process. Thereafter, the calibrated hybrid approach was implemented for forthcoming years. Finally, future land use maps for 2016 and 2026 were predicted by means of this hybrid approach. The simulated maps illustrate a new wave of suburban development in the vicinity of Tehran at the western border of the metropolis during the next decades.

A statistical method for predicting seizure onset zones from human single-neuron recordings

NASA Astrophysics Data System (ADS)

Valdez, André B.; Hickman, Erin N.; Treiman, David M.; Smith, Kris A.; Steinmetz, Peter N.

2013-02-01

Objective. Clinicians often use depth-electrode recordings to localize human epileptogenic foci. To advance the diagnostic value of these recordings, we applied logistic regression models to single-neuron recordings from depth-electrode microwires to predict seizure onset zones (SOZs). Approach. We collected data from 17 epilepsy patients at the Barrow Neurological Institute and developed logistic regression models to calculate the odds of observing SOZs in the hippocampus, amygdala and ventromedial prefrontal cortex, based on statistics such as the burst interspike interval (ISI). Main results. Analysis of these models showed that, for a single-unit increase in burst ISI ratio, the left hippocampus was approximately 12 times more likely to contain a SOZ; and the right amygdala, 14.5 times more likely. Our models were most accurate for the hippocampus bilaterally (at 85% average sensitivity), and performance was comparable with current diagnostics such as electroencephalography. Significance. Logistic regression models can be combined with single-neuron recording to predict likely SOZs in epilepsy patients being evaluated for resective surgery, providing an automated source of clinically useful information.

Hybrid Support Vector Regression and Autoregressive Integrated Moving Average Models Improved by Particle Swarm Optimization for Property Crime Rates Forecasting with Economic Indicators

PubMed Central

Alwee, Razana; Hj Shamsuddin, Siti Mariyam; Sallehuddin, Roselina

2013-01-01

Crimes forecasting is an important area in the field of criminology. Linear models, such as regression and econometric models, are commonly applied in crime forecasting. However, in real crimes data, it is common that the data consists of both linear and nonlinear components. A single model may not be sufficient to identify all the characteristics of the data. The purpose of this study is to introduce a hybrid model that combines support vector regression (SVR) and autoregressive integrated moving average (ARIMA) to be applied in crime rates forecasting. SVR is very robust with small training data and high-dimensional problem. Meanwhile, ARIMA has the ability to model several types of time series. However, the accuracy of the SVR model depends on values of its parameters, while ARIMA is not robust to be applied to small data sets. Therefore, to overcome this problem, particle swarm optimization is used to estimate the parameters of the SVR and ARIMA models. The proposed hybrid model is used to forecast the property crime rates of the United State based on economic indicators. The experimental results show that the proposed hybrid model is able to produce more accurate forecasting results as compared to the individual models. PMID:23766729

Eigenvector Spatial Filtering Regression Modeling of Ground PM2.5 Concentrations Using Remotely Sensed Data.

PubMed

Zhang, Jingyi; Li, Bin; Chen, Yumin; Chen, Meijie; Fang, Tao; Liu, Yongfeng

2018-06-11

This paper proposes a regression model using the Eigenvector Spatial Filtering (ESF) method to estimate ground PM 2.5 concentrations. Covariates are derived from remotely sensed data including aerosol optical depth, normal differential vegetation index, surface temperature, air pressure, relative humidity, height of planetary boundary layer and digital elevation model. In addition, cultural variables such as factory densities and road densities are also used in the model. With the Yangtze River Delta region as the study area, we constructed ESF-based Regression (ESFR) models at different time scales, using data for the period between December 2015 and November 2016. We found that the ESFR models effectively filtered spatial autocorrelation in the OLS residuals and resulted in increases in the goodness-of-fit metrics as well as reductions in residual standard errors and cross-validation errors, compared to the classic OLS models. The annual ESFR model explained 70% of the variability in PM 2.5 concentrations, 16.7% more than the non-spatial OLS model. With the ESFR models, we performed detail analyses on the spatial and temporal distributions of PM 2.5 concentrations in the study area. The model predictions are lower than ground observations but match the general trend. The experiment shows that ESFR provides a promising approach to PM 2.5 analysis and prediction.

Hybrid support vector regression and autoregressive integrated moving average models improved by particle swarm optimization for property crime rates forecasting with economic indicators.

PubMed

Alwee, Razana; Shamsuddin, Siti Mariyam Hj; Sallehuddin, Roselina

2013-01-01

Crimes forecasting is an important area in the field of criminology. Linear models, such as regression and econometric models, are commonly applied in crime forecasting. However, in real crimes data, it is common that the data consists of both linear and nonlinear components. A single model may not be sufficient to identify all the characteristics of the data. The purpose of this study is to introduce a hybrid model that combines support vector regression (SVR) and autoregressive integrated moving average (ARIMA) to be applied in crime rates forecasting. SVR is very robust with small training data and high-dimensional problem. Meanwhile, ARIMA has the ability to model several types of time series. However, the accuracy of the SVR model depends on values of its parameters, while ARIMA is not robust to be applied to small data sets. Therefore, to overcome this problem, particle swarm optimization is used to estimate the parameters of the SVR and ARIMA models. The proposed hybrid model is used to forecast the property crime rates of the United State based on economic indicators. The experimental results show that the proposed hybrid model is able to produce more accurate forecasting results as compared to the individual models.

Evaluation of the Bitterness of Traditional Chinese Medicines using an E-Tongue Coupled with a Robust Partial Least Squares Regression Method.

PubMed

Lin, Zhaozhou; Zhang, Qiao; Liu, Ruixin; Gao, Xiaojie; Zhang, Lu; Kang, Bingya; Shi, Junhan; Wu, Zidan; Gui, Xinjing; Li, Xuelin

2016-01-25

To accurately, safely, and efficiently evaluate the bitterness of Traditional Chinese Medicines (TCMs), a robust predictor was developed using robust partial least squares (RPLS) regression method based on data obtained from an electronic tongue (e-tongue) system. The data quality was verified by the Grubb's test. Moreover, potential outliers were detected based on both the standardized residual and score distance calculated for each sample. The performance of RPLS on the dataset before and after outlier detection was compared to other state-of-the-art methods including multivariate linear regression, least squares support vector machine, and the plain partial least squares regression. Both R² and root-mean-squares error (RMSE) of cross-validation (CV) were recorded for each model. With four latent variables, a robust RMSECV value of 0.3916 with bitterness values ranging from 0.63 to 4.78 were obtained for the RPLS model that was constructed based on the dataset including outliers. Meanwhile, the RMSECV, which was calculated using the models constructed by other methods, was larger than that of the RPLS model. After six outliers were excluded, the performance of all benchmark methods markedly improved, but the difference between the RPLS model constructed before and after outlier exclusion was negligible. In conclusion, the bitterness of TCM decoctions can be accurately evaluated with the RPLS model constructed using e-tongue data.

Scoring and staging systems using cox linear regression modeling and recursive partitioning.

PubMed

Lee, J W; Um, S H; Lee, J B; Mun, J; Cho, H

2006-01-01

Scoring and staging systems are used to determine the order and class of data according to predictors. Systems used for medical data, such as the Child-Turcotte-Pugh scoring and staging systems for ordering and classifying patients with liver disease, are often derived strictly from physicians' experience and intuition. We construct objective and data-based scoring/staging systems using statistical methods. We consider Cox linear regression modeling and recursive partitioning techniques for censored survival data. In particular, to obtain a target number of stages we propose cross-validation and amalgamation algorithms. We also propose an algorithm for constructing scoring and staging systems by integrating local Cox linear regression models into recursive partitioning, so that we can retain the merits of both methods such as superior predictive accuracy, ease of use, and detection of interactions between predictors. The staging system construction algorithms are compared by cross-validation evaluation of real data. The data-based cross-validation comparison shows that Cox linear regression modeling is somewhat better than recursive partitioning when there are only continuous predictors, while recursive partitioning is better when there are significant categorical predictors. The proposed local Cox linear recursive partitioning has better predictive accuracy than Cox linear modeling and simple recursive partitioning. This study indicates that integrating local linear modeling into recursive partitioning can significantly improve prediction accuracy in constructing scoring and staging systems.

Comparison of a Classical and Quantum Based Restricted Boltzmann Machine (RBM) for Application to Non-linear Multivariate Regression.

NASA Astrophysics Data System (ADS)

Dorband, J. E.; Tilak, N.; Radov, A.

2016-12-01

In this paper, a classical computer implementation of RBM is compared to a quantum annealing based RBM running on a D-Wave 2X (an adiabatic quantum computer). The codes for both are essentially identical. Only a flag is set to change the activation function from a classically computed logistic function to the D-Wave. To obtain greater understanding of the behavior of the D-Wave, a study of the stochastic properties of a virtual qubit (a 12 qubit chain) and a cell of qubits (an 8 qubit cell) was performed. We will present the results of comparing the D-Wave implementation with a theoretically errorless adiabatic quantum computer. The main purpose of this study is to develop a generic RBM regression tool in order to infer CO2 fluxes from the NASA satellite OCO-2 observed CO2 concentrations and predicted atmospheric states using regression models. The carbon fluxes will then be assimilated into a land surface model to predict the Net Ecosystem Exchange at globally distributed regional sites.

Comparison between artificial neural network and multilinear regression models in an evaluation of cognitive workload in a flight simulator.

PubMed

Hannula, Manne; Huttunen, Kerttu; Koskelo, Jukka; Laitinen, Tomi; Leino, Tuomo

2008-01-01

In this study, the performances of artificial neural network (ANN) analysis and multilinear regression (MLR) model-based estimation of heart rate were compared in an evaluation of individual cognitive workload. The data comprised electrocardiography (ECG) measurements and an evaluation of cognitive load that induces psychophysiological stress (PPS), collected from 14 interceptor fighter pilots during complex simulated F/A-18 Hornet air battles. In our data, the mean absolute error of the ANN estimate was 11.4 as a visual analog scale score, being 13-23% better than the mean absolute error of the MLR model in the estimation of cognitive workload.

New insights into faster computation of uncertainties

NASA Astrophysics Data System (ADS)

Bhattacharya, Atreyee

2012-11-01

Heavy computation power, lengthy simulations, and an exhaustive number of model runs—often these seem like the only statistical tools that scientists have at their disposal when computing uncertainties associated with predictions, particularly in cases of environmental processes such as groundwater movement. However, calculation of uncertainties need not be as lengthy, a new study shows. Comparing two approaches—the classical Bayesian “credible interval” and a less commonly used regression-based “confidence interval” method—Lu et al. show that for many practical purposes both methods provide similar estimates of uncertainties. The advantage of the regression method is that it demands 10-1000 model runs, whereas the classical Bayesian approach requires 10,000 to millions of model runs.

Estimating suspended sediment load with multivariate adaptive regression spline, teaching-learning based optimization, and artificial bee colony models.

PubMed

Yilmaz, Banu; Aras, Egemen; Nacar, Sinan; Kankal, Murat

2018-05-23

The functional life of a dam is often determined by the rate of sediment delivery to its reservoir. Therefore, an accurate estimate of the sediment load in rivers with dams is essential for designing and predicting a dam's useful lifespan. The most credible method is direct measurements of sediment input, but this can be very costly and it cannot always be implemented at all gauging stations. In this study, we tested various regression models to estimate suspended sediment load (SSL) at two gauging stations on the Çoruh River in Turkey, including artificial bee colony (ABC), teaching-learning-based optimization algorithm (TLBO), and multivariate adaptive regression splines (MARS). These models were also compared with one another and with classical regression analyses (CRA). Streamflow values and previously collected data of SSL were used as model inputs with predicted SSL data as output. Two different training and testing dataset configurations were used to reinforce the model accuracy. For the MARS method, the root mean square error value was found to range between 35% and 39% for the test two gauging stations, which was lower than errors for other models. Error values were even lower (7% to 15%) using another dataset. Our results indicate that simultaneous measurements of streamflow with SSL provide the most effective parameter for obtaining accurate predictive models and that MARS is the most accurate model for predicting SSL. Copyright © 2017 Elsevier B.V. All rights reserved.

The application of neural network model to the simulation nitrous oxide emission in the hydro-fluctuation belt of Three Gorges Reservoir

NASA Astrophysics Data System (ADS)

Song, Lanlan

2017-04-01

Nitrous oxide is much more potent greenhouse gas than carbon dioxide. However, the estimation of N2O flux is usually clouded with uncertainty, mainly due to high spatial and temporal variations. This hampers the development of general mechanistic models for N2O emission as well, as most previously developed models were empirical or exhibited low predictability with numerous assumptions. In this study, we tested General Regression Neural Networks (GRNN) as an alternative to classic empirical models for simulating N2O emission in riparian zones of Reservoirs. GRNN and nonlinear regression (NLR) were applied to estimate the N2O flux of 1-year observations in riparian zones of Three Gorge Reservoir. NLR resulted in lower prediction power and higher residuals compared to GRNN. Although nonlinear regression model estimated similar average values of N2O, it could not capture the fluctuation patterns accurately. In contrast, GRNN model achieved a fairly high predictability, with an R2 of 0.59 for model validation, 0.77 for model calibration (training), and a low root mean square error (RMSE), indicating a high capacity to simulate the dynamics of N2O flux. According to a sensitivity analysis of the GRNN, nonlinear relationships between input variables and N2O flux were well explained. Our results suggest that the GRNN developed in this study has a greater performance in simulating variations in N2O flux than nonlinear regressions.

Weather Impact on Airport Arrival Meter Fix Throughput

NASA Technical Reports Server (NTRS)

Wang, Yao

2017-01-01

Time-based flow management provides arrival aircraft schedules based on arrival airport conditions, airport capacity, required spacing, and weather conditions. In order to meet a scheduled time at which arrival aircraft can cross an airport arrival meter fix prior to entering the airport terminal airspace, air traffic controllers make regulations on air traffic. Severe weather may create an airport arrival bottleneck if one or more of airport arrival meter fixes are partially or completely blocked by the weather and the arrival demand has not been reduced accordingly. Under these conditions, aircraft are frequently being put in holding patterns until they can be rerouted. A model that predicts the weather impacted meter fix throughput may help air traffic controllers direct arrival flows into the airport more efficiently, minimizing arrival meter fix congestion. This paper presents an analysis of air traffic flows across arrival meter fixes at the Newark Liberty International Airport (EWR). Several scenarios of weather impacted EWR arrival fix flows are described. Furthermore, multiple linear regression and regression tree ensemble learning approaches for translating multiple sector Weather Impacted Traffic Indexes (WITI) to EWR arrival meter fix throughputs are examined. These weather translation models are developed and validated using the EWR arrival flight and weather data for the period of April-September in 2014. This study also compares the performance of the regression tree ensemble with traditional multiple linear regression models for estimating the weather impacted throughputs at each of the EWR arrival meter fixes. For all meter fixes investigated, the results from the regression tree ensemble weather translation models show a stronger correlation between model outputs and observed meter fix throughputs than that produced from multiple linear regression method.

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.

Performance of the modified Poisson regression approach for estimating relative risks from clustered prospective data.

PubMed

Yelland, Lisa N; Salter, Amy B; Ryan, Philip

2011-10-15

Modified Poisson regression, which combines a log Poisson regression model with robust variance estimation, is a useful alternative to log binomial regression for estimating relative risks. Previous studies have shown both analytically and by simulation that modified Poisson regression is appropriate for independent prospective data. This method is often applied to clustered prospective data, despite a lack of evidence to support its use in this setting. The purpose of this article is to evaluate the performance of the modified Poisson regression approach for estimating relative risks from clustered prospective data, by using generalized estimating equations to account for clustering. A simulation study is conducted to compare log binomial regression and modified Poisson regression for analyzing clustered data from intervention and observational studies. Both methods generally perform well in terms of bias, type I error, and coverage. Unlike log binomial regression, modified Poisson regression is not prone to convergence problems. The methods are contrasted by using example data sets from 2 large studies. The results presented in this article support the use of modified Poisson regression as an alternative to log binomial regression for analyzing clustered prospective data when clustering is taken into account by using generalized estimating equations.

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.

Genomic Bayesian functional regression models with interactions for predicting wheat grain yield using hyper-spectral image data.

PubMed

Montesinos-López, Abelardo; Montesinos-López, Osval A; Cuevas, Jaime; Mata-López, Walter A; Burgueño, Juan; Mondal, Sushismita; Huerta, Julio; Singh, Ravi; Autrique, Enrique; González-Pérez, Lorena; Crossa, José

2017-01-01

Modern agriculture uses hyperspectral cameras that provide hundreds of reflectance data at discrete narrow bands in many environments. These bands often cover the whole visible light spectrum and part of the infrared and ultraviolet light spectra. With the bands, vegetation indices are constructed for predicting agronomically important traits such as grain yield and biomass. However, since vegetation indices only use some wavelengths (referred to as bands), we propose using all bands simultaneously as predictor variables for the primary trait grain yield; results of several multi-environment maize (Aguate et al. in Crop Sci 57(5):1-8, 2017) and wheat (Montesinos-López et al. in Plant Methods 13(4):1-23, 2017) breeding trials indicated that using all bands produced better prediction accuracy than vegetation indices. However, until now, these prediction models have not accounted for the effects of genotype × environment (G × E) and band × environment (B × E) interactions incorporating genomic or pedigree information. In this study, we propose Bayesian functional regression models that take into account all available bands, genomic or pedigree information, the main effects of lines and environments, as well as G × E and B × E interaction effects. The data set used is comprised of 976 wheat lines evaluated for grain yield in three environments (Drought, Irrigated and Reduced Irrigation). The reflectance data were measured in 250 discrete narrow bands ranging from 392 to 851 nm (nm). The proposed Bayesian functional regression models were implemented using two types of basis: B-splines and Fourier. Results of the proposed Bayesian functional regression models, including all the wavelengths for predicting grain yield, were compared with results from conventional models with and without bands. We observed that the models with B × E interaction terms were the most accurate models, whereas the functional regression models (with B-splines and Fourier basis) and the conventional models performed similarly in terms of prediction accuracy. However, the functional regression models are more parsimonious and computationally more efficient because the number of beta coefficients to be estimated is 21 (number of basis), rather than estimating the 250 regression coefficients for all bands. In this study adding pedigree or genomic information did not increase prediction accuracy.

Analytics of Radioactive Materials Released in the Fukushima Daiichi Nuclear Accident

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

Egarievwe, Stephen U.; Nuclear Engineering Department, University of Tennessee, Knoxville, TN; Coble, Jamie B.

The 2011 Fukushima Daiichi nuclear accident in Japan resulted in the release of radioactive materials into the atmosphere, the nearby sea, and the surrounding land. Following the accident, several meteorological models were used to predict the transport of the radioactive materials to other continents such as North America and Europe. Also of high importance is the dispersion of radioactive materials locally and within Japan. Based on the International Atomic Energy Agency (IAEA) Convention on Early Notification of a nuclear accident, several radiological data sets were collected on the accident by the Japanese authorities. Among the radioactive materials monitored, are I-131more » and Cs-137 which form the major contributions to the contamination of drinking water. The radiation dose in the atmosphere was also measured. It is impractical to measure contamination and radiation dose in every place of interest. Therefore, modeling helps to predict contamination and radiation dose. Some modeling studies that have been reported in the literature include the simulation of transport and deposition of I-131 and Cs-137 from the accident, Cs-137 deposition and contamination of Japanese soils, and preliminary estimates of I-131 and Cs-137 discharged from the plant into the atmosphere. In this paper, we present statistical analytics of I-131 and Cs-137 with the goal of predicting gamma dose from the Fukushima Daiichi nuclear accident. The data sets used in our study were collected from the IAEA Fukushima Monitoring Database. As part of this study, we investigated several regression models to find the best algorithm for modeling the gamma dose. The modeling techniques used in our study include linear regression, principal component regression (PCR), partial least square (PLS) regression, and ridge regression. Our preliminary results on the first set of data showed that the linear regression model with one variable was the best with a root mean square error of 0.0133 μSv/h, compared to 0.0210 μSv/h for PCR, 0.231 μSv/h for ridge regression L-curve, 0.0856 μSv/h for PLS, and 0.0860 μSv/h for ridge regression cross validation. Complete results using the full datasets for these models will also be presented. (authors)« less

Hybrid rocket engine, theoretical model and experiment

NASA Astrophysics Data System (ADS)

Chelaru, Teodor-Viorel; Mingireanu, Florin

2011-06-01

The purpose of this paper is to build a theoretical model for the hybrid rocket engine/motor and to validate it using experimental results. The work approaches the main problems of the hybrid motor: the scalability, the stability/controllability of the operating parameters and the increasing of the solid fuel regression rate. At first, we focus on theoretical models for hybrid rocket motor and compare the results with already available experimental data from various research groups. A primary computation model is presented together with results from a numerical algorithm based on a computational model. We present theoretical predictions for several commercial hybrid rocket motors, having different scales and compare them with experimental measurements of those hybrid rocket motors. Next the paper focuses on tribrid rocket motor concept, which by supplementary liquid fuel injection can improve the thrust controllability. A complementary computation model is also presented to estimate regression rate increase of solid fuel doped with oxidizer. Finally, the stability of the hybrid rocket motor is investigated using Liapunov theory. Stability coefficients obtained are dependent on burning parameters while the stability and command matrixes are identified. The paper presents thoroughly the input data of the model, which ensures the reproducibility of the numerical results by independent researchers.

USE OF WEIBULL FUNCTION FOR NON-LINEAR ANALYSIS OF EFFECTS OF LOW LEVELS OF SIMULATED HERBICIDE DRIFT ON PLANTS

EPA Science Inventory

We compared two regression models, which are based on the Weibull and probit functions, for the analysis of pesticide toxicity data from laboratory studies on Illinois crop and native plant species. Both mathematical models are continuous, differentiable, strictly positive, and...

Artificial Neural Networks: A New Approach to Predicting Application Behavior.

ERIC Educational Resources Information Center

Gonzalez, Julie M. Byers; DesJardins, Stephen L.

2002-01-01

Applied the technique of artificial neural networks to predict which students were likely to apply to one research university. Compared the results to the traditional analysis tool, logistic regression modeling. Found that the addition of artificial intelligence models was a useful new tool for predicting student application behavior. (EV)

Stochastic search, optimization and regression with energy applications

NASA Astrophysics Data System (ADS)

Hannah, Lauren A.

Designing clean energy systems will be an important task over the next few decades. One of the major roadblocks is a lack of mathematical tools to economically evaluate those energy systems. However, solutions to these mathematical problems are also of interest to the operations research and statistical communities in general. This thesis studies three problems that are of interest to the energy community itself or provide support for solution methods: R&D portfolio optimization, nonparametric regression and stochastic search with an observable state variable. First, we consider the one stage R&D portfolio optimization problem to avoid the sequential decision process associated with the multi-stage. The one stage problem is still difficult because of a non-convex, combinatorial decision space and a non-convex objective function. We propose a heuristic solution method that uses marginal project values---which depend on the selected portfolio---to create a linear objective function. In conjunction with the 0-1 decision space, this new problem can be solved as a knapsack linear program. This method scales well to large decision spaces. We also propose an alternate, provably convergent algorithm that does not exploit problem structure. These methods are compared on a solid oxide fuel cell R&D portfolio problem. Next, we propose Dirichlet Process mixtures of Generalized Linear Models (DPGLM), a new method of nonparametric regression that accommodates continuous and categorical inputs, and responses that can be modeled by a generalized linear model. We prove conditions for the asymptotic unbiasedness of the DP-GLM regression mean function estimate. We also give examples for when those conditions hold, including models for compactly supported continuous distributions and a model with continuous covariates and categorical response. We empirically analyze the properties of the DP-GLM and why it provides better results than existing Dirichlet process mixture regression models. We evaluate DP-GLM on several data sets, comparing it to modern methods of nonparametric regression like CART, Bayesian trees and Gaussian processes. Compared to existing techniques, the DP-GLM provides a single model (and corresponding inference algorithms) that performs well in many regression settings. Finally, we study convex stochastic search problems where a noisy objective function value is observed after a decision is made. There are many stochastic search problems whose behavior depends on an exogenous state variable which affects the shape of the objective function. Currently, there is no general purpose algorithm to solve this class of problems. We use nonparametric density estimation to take observations from the joint state-outcome distribution and use them to infer the optimal decision for a given query state. We propose two solution methods that depend on the problem characteristics: function-based and gradient-based optimization. We examine two weighting schemes, kernel-based weights and Dirichlet process-based weights, for use with the solution methods. The weights and solution methods are tested on a synthetic multi-product newsvendor problem and the hour-ahead wind commitment problem. Our results show that in some cases Dirichlet process weights offer substantial benefits over kernel based weights and more generally that nonparametric estimation methods provide good solutions to otherwise intractable problems.

Cost-of-illness studies based on massive data: a prevalence-based, top-down regression approach.

PubMed

Stollenwerk, Björn; Welchowski, Thomas; Vogl, Matthias; Stock, Stephanie

2016-04-01

Despite the increasing availability of routine data, no analysis method has yet been presented for cost-of-illness (COI) studies based on massive data. We aim, first, to present such a method and, second, to assess the relevance of the associated gain in numerical efficiency. We propose a prevalence-based, top-down regression approach consisting of five steps: aggregating the data; fitting a generalized additive model (GAM); predicting costs via the fitted GAM; comparing predicted costs between prevalent and non-prevalent subjects; and quantifying the stochastic uncertainty via error propagation. To demonstrate the method, it was applied to aggregated data in the context of chronic lung disease to German sickness funds data (from 1999), covering over 7.3 million insured. To assess the gain in numerical efficiency, the computational time of the innovative approach has been compared with corresponding GAMs applied to simulated individual-level data. Furthermore, the probability of model failure was modeled via logistic regression. Applying the innovative method was reasonably fast (19 min). In contrast, regarding patient-level data, computational time increased disproportionately by sample size. Furthermore, using patient-level data was accompanied by a substantial risk of model failure (about 80 % for 6 million subjects). The gain in computational efficiency of the innovative COI method seems to be of practical relevance. Furthermore, it may yield more precise cost estimates.

An hourly PM10 diagnosis model for the Bilbao metropolitan area using a linear regression methodology.

PubMed

González-Aparicio, I; Hidalgo, J; Baklanov, A; Padró, A; Santa-Coloma, O

2013-07-01

There is extensive evidence of the negative impacts on health linked to the rise of the regional background of particulate matter (PM) 10 levels. These levels are often increased over urban areas becoming one of the main air pollution concerns. This is the case on the Bilbao metropolitan area, Spain. This study describes a data-driven model to diagnose PM10 levels in Bilbao at hourly intervals. The model is built with a training period of 7-year historical data covering different urban environments (inland, city centre and coastal sites). The explanatory variables are quantitative-log [NO2], temperature, short-wave incoming radiation, wind speed and direction, specific humidity, hour and vehicle intensity-and qualitative-working days/weekends, season (winter/summer), the hour (from 00 to 23 UTC) and precipitation/no precipitation. Three different linear regression models are compared: simple linear regression; linear regression with interaction terms (INT); and linear regression with interaction terms following the Sawa's Bayesian Information Criteria (INT-BIC). Each type of model is calculated selecting two different periods: the training (it consists of 6 years) and the testing dataset (it consists of 1 year). The results of each type of model show that the INT-BIC-based model (R(2) = 0.42) is the best. Results were R of 0.65, 0.63 and 0.60 for the city centre, inland and coastal sites, respectively, a level of confidence similar to the state-of-the art methodology. The related error calculated for longer time intervals (monthly or seasonal means) diminished significantly (R of 0.75-0.80 for monthly means and R of 0.80 to 0.98 at seasonally means) with respect to shorter periods.

Multivariate Bias Correction Procedures for Improving Water Quality Predictions from the SWAT Model

NASA Astrophysics Data System (ADS)

Arumugam, S.; Libera, D.

2017-12-01

Water quality observations are usually not available on a continuous basis for longer than 1-2 years at a time over a decadal period given the labor requirements making calibrating and validating mechanistic models difficult. Further, any physical model predictions inherently have bias (i.e., under/over estimation) and require post-simulation techniques to preserve the long-term mean monthly attributes. This study suggests a multivariate bias-correction technique and compares to a common technique in improving the performance of the SWAT model in predicting daily streamflow and TN loads across the southeast based on split-sample validation. The approach is a dimension reduction technique, canonical correlation analysis (CCA) that regresses the observed multivariate attributes with the SWAT model simulated values. The common approach is a regression based technique that uses an ordinary least squares regression to adjust model values. The observed cross-correlation between loadings and streamflow is better preserved when using canonical correlation while simultaneously reducing individual biases. Additionally, canonical correlation analysis does a better job in preserving the observed joint likelihood of observed streamflow and loadings. These procedures were applied to 3 watersheds chosen from the Water Quality Network in the Southeast Region; specifically, watersheds with sufficiently large drainage areas and number of observed data points. The performance of these two approaches are compared for the observed period and over a multi-decadal period using loading estimates from the USGS LOADEST model. Lastly, the CCA technique is applied in a forecasting sense by using 1-month ahead forecasts of P & T from ECHAM4.5 as forcings in the SWAT model. Skill in using the SWAT model for forecasting loadings and streamflow at the monthly and seasonal timescale is also discussed.

Feature Extraction of Event-Related Potentials Using Wavelets: An Application to Human Performance Monitoring

NASA Technical Reports Server (NTRS)

Trejo, Leonard J.; Shensa, Mark J.; Remington, Roger W. (Technical Monitor)

1998-01-01

This report describes the development and evaluation of mathematical models for predicting human performance from discrete wavelet transforms (DWT) of event-related potentials (ERP) elicited by task-relevant stimuli. The DWT was compared to principal components analysis (PCA) for representation of ERPs in linear regression and neural network models developed to predict a composite measure of human signal detection performance. Linear regression models based on coefficients of the decimated DWT predicted signal detection performance with half as many f ree parameters as comparable models based on PCA scores. In addition, the DWT-based models were more resistant to model degradation due to over-fitting than PCA-based models. Feed-forward neural networks were trained using the backpropagation,-, algorithm to predict signal detection performance based on raw ERPs, PCA scores, or high-power coefficients of the DWT. Neural networks based on high-power DWT coefficients trained with fewer iterations, generalized to new data better, and were more resistant to overfitting than networks based on raw ERPs. Networks based on PCA scores did not generalize to new data as well as either the DWT network or the raw ERP network. The results show that wavelet expansions represent the ERP efficiently and extract behaviorally important features for use in linear regression or neural network models of human performance. The efficiency of the DWT is discussed in terms of its decorrelation and energy compaction properties. In addition, the DWT models provided evidence that a pattern of low-frequency activity (1 to 3.5 Hz) occurring at specific times and scalp locations is a reliable correlate of human signal detection performance.

Feature extraction of event-related potentials using wavelets: an application to human performance monitoring

NASA Technical Reports Server (NTRS)

Trejo, L. J.; Shensa, M. J.

1999-01-01

This report describes the development and evaluation of mathematical models for predicting human performance from discrete wavelet transforms (DWT) of event-related potentials (ERP) elicited by task-relevant stimuli. The DWT was compared to principal components analysis (PCA) for representation of ERPs in linear regression and neural network models developed to predict a composite measure of human signal detection performance. Linear regression models based on coefficients of the decimated DWT predicted signal detection performance with half as many free parameters as comparable models based on PCA scores. In addition, the DWT-based models were more resistant to model degradation due to over-fitting than PCA-based models. Feed-forward neural networks were trained using the backpropagation algorithm to predict signal detection performance based on raw ERPs, PCA scores, or high-power coefficients of the DWT. Neural networks based on high-power DWT coefficients trained with fewer iterations, generalized to new data better, and were more resistant to overfitting than networks based on raw ERPs. Networks based on PCA scores did not generalize to new data as well as either the DWT network or the raw ERP network. The results show that wavelet expansions represent the ERP efficiently and extract behaviorally important features for use in linear regression or neural network models of human performance. The efficiency of the DWT is discussed in terms of its decorrelation and energy compaction properties. In addition, the DWT models provided evidence that a pattern of low-frequency activity (1 to 3.5 Hz) occurring at specific times and scalp locations is a reliable correlate of human signal detection performance. Copyright 1999 Academic Press.

Comparing methods of analysing datasets with small clusters: case studies using four paediatric datasets.

PubMed

Marston, Louise; Peacock, Janet L; Yu, Keming; Brocklehurst, Peter; Calvert, Sandra A; Greenough, Anne; Marlow, Neil

2009-07-01

Studies of prematurely born infants contain a relatively large percentage of multiple births, so the resulting data have a hierarchical structure with small clusters of size 1, 2 or 3. Ignoring the clustering may lead to incorrect inferences. The aim of this study was to compare statistical methods which can be used to analyse such data: generalised estimating equations, multilevel models, multiple linear regression and logistic regression. Four datasets which differed in total size and in percentage of multiple births (n = 254, multiple 18%; n = 176, multiple 9%; n = 10 098, multiple 3%; n = 1585, multiple 8%) were analysed. With the continuous outcome, two-level models produced similar results in the larger dataset, while generalised least squares multilevel modelling (ML GLS 'xtreg' in Stata) and maximum likelihood multilevel modelling (ML MLE 'xtmixed' in Stata) produced divergent estimates using the smaller dataset. For the dichotomous outcome, most methods, except generalised least squares multilevel modelling (ML GH 'xtlogit' in Stata) gave similar odds ratios and 95% confidence intervals within datasets. For the continuous outcome, our results suggest using multilevel modelling. We conclude that generalised least squares multilevel modelling (ML GLS 'xtreg' in Stata) and maximum likelihood multilevel modelling (ML MLE 'xtmixed' in Stata) should be used with caution when the dataset is small. Where the outcome is dichotomous and there is a relatively large percentage of non-independent data, it is recommended that these are accounted for in analyses using logistic regression with adjusted standard errors or multilevel modelling. If, however, the dataset has a small percentage of clusters greater than size 1 (e.g. a population dataset of children where there are few multiples) there appears to be less need to adjust for clustering.

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.

Binary Logistic Regression Versus Boosted Regression Trees in Assessing Landslide Susceptibility for Multiple-Occurring Regional Landslide Events: Application to the 2009 Storm Event in Messina (Sicily, southern Italy).

NASA Astrophysics Data System (ADS)

Lombardo, L.; Cama, M.; Maerker, M.; Parisi, L.; Rotigliano, E.

2014-12-01

This study aims at comparing the performances of Binary Logistic Regression (BLR) and Boosted Regression Trees (BRT) methods in assessing landslide susceptibility for multiple-occurrence regional landslide events within the Mediterranean region. A test area was selected in the north-eastern sector of Sicily (southern Italy), corresponding to the catchments of the Briga and the Giampilieri streams both stretching for few kilometres from the Peloritan ridge (eastern Sicily, Italy) to the Ionian sea. This area was struck on the 1st October 2009 by an extreme climatic event resulting in thousands of rapid shallow landslides, mainly of debris flows and debris avalanches types involving the weathered layer of a low to high grade metamorphic bedrock. Exploiting the same set of predictors and the 2009 landslide archive, BLR- and BRT-based susceptibility models were obtained for the two catchments separately, adopting a random partition (RP) technique for validation; besides, the models trained in one of the two catchments (Briga) were tested in predicting the landslide distribution in the other (Giampilieri), adopting a spatial partition (SP) based validation procedure. All the validation procedures were based on multi-folds tests so to evaluate and compare the reliability of the fitting, the prediction skill, the coherence in the predictor selection and the precision of the susceptibility estimates. All the obtained models for the two methods produced very high predictive performances, with a general congruence between BLR and BRT in the predictor importance. In particular, the research highlighted that BRT-models reached a higher prediction performance with respect to BLR-models, for RP based modelling, whilst for the SP-based models the difference in predictive skills between the two methods dropped drastically, converging to an analogous excellent performance. However, when looking at the precision of the probability estimates, BLR demonstrated to produce more robust models in terms of selected predictors and coefficients, as well as of dispersion of the estimated probabilities around the mean value for each mapped pixel. The difference in the behaviour could be interpreted as the result of overfitting effects, which heavily affect decision tree classification more than logistic regression techniques.

Logistic LASSO regression for the diagnosis of breast cancer using clinical demographic data and the BI-RADS lexicon for ultrasonography.

PubMed

Kim, Sun Mi; Kim, Yongdai; Jeong, Kuhwan; Jeong, Heeyeong; Kim, Jiyoung

2018-01-01

The aim of this study was to compare the performance of image analysis for predicting breast cancer using two distinct regression models and to evaluate the usefulness of incorporating clinical and demographic data (CDD) into the image analysis in order to improve the diagnosis of breast cancer. This study included 139 solid masses from 139 patients who underwent a ultrasonography-guided core biopsy and had available CDD between June 2009 and April 2010. Three breast radiologists retrospectively reviewed 139 breast masses and described each lesion using the Breast Imaging Reporting and Data System (BI-RADS) lexicon. We applied and compared two regression methods-stepwise logistic (SL) regression and logistic least absolute shrinkage and selection operator (LASSO) regression-in which the BI-RADS descriptors and CDD were used as covariates. We investigated the performances of these regression methods and the agreement of radiologists in terms of test misclassification error and the area under the curve (AUC) of the tests. Logistic LASSO regression was superior (P<0.05) to SL regression, regardless of whether CDD was included in the covariates, in terms of test misclassification errors (0.234 vs. 0.253, without CDD; 0.196 vs. 0.258, with CDD) and AUC (0.785 vs. 0.759, without CDD; 0.873 vs. 0.735, with CDD). However, it was inferior (P<0.05) to the agreement of three radiologists in terms of test misclassification errors (0.234 vs. 0.168, without CDD; 0.196 vs. 0.088, with CDD) and the AUC without CDD (0.785 vs. 0.844, P<0.001), but was comparable to the AUC with CDD (0.873 vs. 0.880, P=0.141). Logistic LASSO regression based on BI-RADS descriptors and CDD showed better performance than SL in predicting the presence of breast cancer. The use of CDD as a supplement to the BI-RADS descriptors significantly improved the prediction of breast cancer using logistic LASSO regression.

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.

Plant leaf chlorophyll content retrieval based on a field imaging spectroscopy system.

PubMed

Liu, Bo; Yue, Yue-Min; Li, Ru; Shen, Wen-Jing; Wang, Ke-Lin

2014-10-23

A field imaging spectrometer system (FISS; 380-870 nm and 344 bands) was designed for agriculture applications. In this study, FISS was used to gather spectral information from soybean leaves. The chlorophyll content was retrieved using a multiple linear regression (MLR), partial least squares (PLS) regression and support vector machine (SVM) regression. Our objective was to verify the performance of FISS in a quantitative spectral analysis through the estimation of chlorophyll content and to determine a proper quantitative spectral analysis method for processing FISS data. The results revealed that the derivative reflectance was a more sensitive indicator of chlorophyll content and could extract content information more efficiently than the spectral reflectance, which is more significant for FISS data compared to ASD (analytical spectral devices) data, reducing the corresponding RMSE (root mean squared error) by 3.3%-35.6%. Compared with the spectral features, the regression methods had smaller effects on the retrieval accuracy. A multivariate linear model could be the ideal model to retrieve chlorophyll information with a small number of significant wavelengths used. The smallest RMSE of the chlorophyll content retrieved using FISS data was 0.201 mg/g, a relative reduction of more than 30% compared with the RMSE based on a non-imaging ASD spectrometer, which represents a high estimation accuracy compared with the mean chlorophyll content of the sampled leaves (4.05 mg/g). Our study indicates that FISS could obtain both spectral and spatial detailed information of high quality. Its image-spectrum-in-one merit promotes the good performance of FISS in quantitative spectral analyses, and it can potentially be widely used in the agricultural sector.

Plant Leaf Chlorophyll Content Retrieval Based on a Field Imaging Spectroscopy System

PubMed Central

Liu, Bo; Yue, Yue-Min; Li, Ru; Shen, Wen-Jing; Wang, Ke-Lin

2014-01-01

A field imaging spectrometer system (FISS; 380–870 nm and 344 bands) was designed for agriculture applications. In this study, FISS was used to gather spectral information from soybean leaves. The chlorophyll content was retrieved using a multiple linear regression (MLR), partial least squares (PLS) regression and support vector machine (SVM) regression. Our objective was to verify the performance of FISS in a quantitative spectral analysis through the estimation of chlorophyll content and to determine a proper quantitative spectral analysis method for processing FISS data. The results revealed that the derivative reflectance was a more sensitive indicator of chlorophyll content and could extract content information more efficiently than the spectral reflectance, which is more significant for FISS data compared to ASD (analytical spectral devices) data, reducing the corresponding RMSE (root mean squared error) by 3.3%–35.6%. Compared with the spectral features, the regression methods had smaller effects on the retrieval accuracy. A multivariate linear model could be the ideal model to retrieve chlorophyll information with a small number of significant wavelengths used. The smallest RMSE of the chlorophyll content retrieved using FISS data was 0.201 mg/g, a relative reduction of more than 30% compared with the RMSE based on a non-imaging ASD spectrometer, which represents a high estimation accuracy compared with the mean chlorophyll content of the sampled leaves (4.05 mg/g). Our study indicates that FISS could obtain both spectral and spatial detailed information of high quality. Its image-spectrum-in-one merit promotes the good performance of FISS in quantitative spectral analyses, and it can potentially be widely used in the agricultural sector. PMID:25341439

Freedom Solo Versus Trifecta Bioprotheses: Clinical and Hemodynamic Evaluation after Propensity Score Matching.

PubMed

J Cerqueira, Rui; Melo, Renata; Moreira, Soraia; A Saraiva, Francisca; Andrade, Marta; Salgueiro, Elson; Almeida, Jorge; J Amorim, Mário; Pinho, Paulo; Lourenço, André; F Leite-Moreira, Adelino

2017-01-01

To compare stentless Freedom Solo and stented Trifecta aortic bioprostheses regarding hemodynamic profile, left ventricular mass regression, early and late postoperative outcomes and survival. Longitudinal cohort study of consecutive patients undergoing aortic valve replacement (from 2009 to 2016) with either Freedom Solo or Trifecta at one centre. Local databases and national records were queried. Postoperative echocardiography (3-6 months) was obtained for hemodynamic profile (mean transprosthetic gradient and effective orifice area) and left ventricle mass determination. After propensity score matching (21 covariates), Kaplan-Meier analysis and cumulative incidence analysis were performed for survival and combined outcome of structural valve deterioration and endocarditis, respectively. Hemodynamics and left ventricle mass regression were assessed by a mixed- -effects model including propensity score as a covariate. From a total sample of 397 Freedom Solo and 525 Trifecta patients with a median follow-up time of 4.0 (2.2- 6.0) and 2.4 (1.4-3.7) years, respectively, a matched sample of 329 pairs was obtained. Well-balanced matched groups showed no difference in survival (hazard ratio=1.04, 95% confidence interval=0.69-1.56) or cumulative hazards of combined outcome (subhazard ratio=0.54, 95% confidence interval=0.21-1.39). Although Trifecta showed improved hemodynamic profile compared to Freedom Solo, no differences were found in left ventricle mass regression. Trifecta has a slightly improved hemodynamic profile compared to Freedom Solo but this does not translate into differences in the extent of mass regression, postoperative outcomes or survival, which were good and comparable for both bioprostheses. Long-term follow-up is needed for comparisons with older models of bioprostheses.

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.

A self-trained classification technique for producing 30 m percent-water maps from Landsat data

USGS Publications Warehouse

Rover, Jennifer R.; Wylie, Bruce K.; Ji, Lei

2010-01-01

Small bodies of water can be mapped with moderate-resolution satellite data using methods where water is mapped as subpixel fractions using field measurements or high-resolution images as training datasets. A new method, developed from a regression-tree technique, uses a 30 m Landsat image for training the regression tree that, in turn, is applied to the same image to map subpixel water. The self-trained method was evaluated by comparing the percent-water map with three other maps generated from established percent-water mapping methods: (1) a regression-tree model trained with a 5 m SPOT 5 image, (2) a regression-tree model based on endmembers and (3) a linear unmixing classification technique. The results suggest that subpixel water fractions can be accurately estimated when high-resolution satellite data or intensively interpreted training datasets are not available, which increases our ability to map small water bodies or small changes in lake size at a regional scale.

Assessment and improvement of biotransfer models to cow's milk and beef used in exposure assessment tools for organic pollutants.

PubMed

Takaki, Koki; Wade, Andrew J; Collins, Chris D

2015-11-01

The aim of this study was to assess and improve the accuracy of biotransfer models for the organic pollutants (PCBs, PCDD/Fs, PBDEs, PFCAs, and pesticides) into cow's milk and beef used in human exposure assessment. Metabolic rate in cattle is known as a key parameter for this biotransfer, however few experimental data and no simulation methods are currently available. In this research, metabolic rate was estimated using existing QSAR biodegradation models of microorganisms (BioWIN) and fish (EPI-HL and IFS-HL). This simulated metabolic rate was then incorporated into the mechanistic cattle biotransfer models (RAIDAR, ACC-HUMAN, OMEGA, and CKow). The goodness of fit tests showed that RAIDAR, ACC-HUMAN, OMEGA model performances were significantly improved using either of the QSARs when comparing the new model outputs to observed data. The CKow model is the only one that separates the processes in the gut and liver. This model showed the lowest residual error of all the models tested when the BioWIN model was used to represent the ruminant metabolic process in the gut and the two fish QSARs were used to represent the metabolic process in the liver. Our testing included EUSES and CalTOX which are KOW-regression models that are widely used in regulatory assessment. New regressions based on the simulated rate of the two metabolic processes are also proposed as an alternative to KOW-regression models for a screening risk assessment. The modified CKow model is more physiologically realistic, but has equivalent usability to existing KOW-regression models for estimating cattle biotransfer of organic pollutants. Copyright © 2015. Published by Elsevier Ltd.

Spatial variability of excess mortality during prolonged dust events in a high-density city: a time-stratified spatial regression approach.

PubMed

Wong, Man Sing; Ho, Hung Chak; Yang, Lin; Shi, Wenzhong; Yang, Jinxin; Chan, Ta-Chien

2017-07-24

Dust events have long been recognized to be associated with a higher mortality risk. However, no study has investigated how prolonged dust events affect the spatial variability of mortality across districts in a downwind city. In this study, we applied a spatial regression approach to estimate the district-level mortality during two extreme dust events in Hong Kong. We compared spatial and non-spatial models to evaluate the ability of each regression to estimate mortality. We also compared prolonged dust events with non-dust events to determine the influences of community factors on mortality across the city. The density of a built environment (estimated by the sky view factor) had positive association with excess mortality in each district, while socioeconomic deprivation contributed by lower income and lower education induced higher mortality impact in each territory planning unit during a prolonged dust event. Based on the model comparison, spatial error modelling with the 1st order of queen contiguity consistently outperformed other models. The high-risk areas with higher increase in mortality were located in an urban high-density environment with higher socioeconomic deprivation. Our model design shows the ability to predict spatial variability of mortality risk during an extreme weather event that is not able to be estimated based on traditional time-series analysis or ecological studies. Our spatial protocol can be used for public health surveillance, sustainable planning and disaster preparation when relevant data are available.

Time Series Analysis for Forecasting Hospital Census: Application to the Neonatal Intensive Care Unit

PubMed Central

Hoover, Stephen; Jackson, Eric V.; Paul, David; Locke, Robert

2016-01-01

Summary Background Accurate prediction of future patient census in hospital units is essential for patient safety, health outcomes, and resource planning. Forecasting census in the Neonatal Intensive Care Unit (NICU) is particularly challenging due to limited ability to control the census and clinical trajectories. The fixed average census approach, using average census from previous year, is a forecasting alternative used in clinical practice, but has limitations due to census variations. Objective Our objectives are to: (i) analyze the daily NICU census at a single health care facility and develop census forecasting models, (ii) explore models with and without patient data characteristics obtained at the time of admission, and (iii) evaluate accuracy of the models compared with the fixed average census approach. Methods We used five years of retrospective daily NICU census data for model development (January 2008 – December 2012, N=1827 observations) and one year of data for validation (January – December 2013, N=365 observations). Best-fitting models of ARIMA and linear regression were applied to various 7-day prediction periods and compared using error statistics. Results The census showed a slightly increasing linear trend. Best fitting models included a non-seasonal model, ARIMA(1,0,0), seasonal ARIMA models, ARIMA(1,0,0)x(1,1,2)7 and ARIMA(2,1,4)x(1,1,2)14, as well as a seasonal linear regression model. Proposed forecasting models resulted on average in 36.49% improvement in forecasting accuracy compared with the fixed average census approach. Conclusions Time series models provide higher prediction accuracy under different census conditions compared with the fixed average census approach. Presented methodology is easily applicable in clinical practice, can be generalized to other care settings, support short- and long-term census forecasting, and inform staff resource planning. PMID:27437040

Time Series Analysis for Forecasting Hospital Census: Application to the Neonatal Intensive Care Unit.

PubMed

Capan, Muge; Hoover, Stephen; Jackson, Eric V; Paul, David; Locke, Robert

2016-01-01

Accurate prediction of future patient census in hospital units is essential for patient safety, health outcomes, and resource planning. Forecasting census in the Neonatal Intensive Care Unit (NICU) is particularly challenging due to limited ability to control the census and clinical trajectories. The fixed average census approach, using average census from previous year, is a forecasting alternative used in clinical practice, but has limitations due to census variations. Our objectives are to: (i) analyze the daily NICU census at a single health care facility and develop census forecasting models, (ii) explore models with and without patient data characteristics obtained at the time of admission, and (iii) evaluate accuracy of the models compared with the fixed average census approach. We used five years of retrospective daily NICU census data for model development (January 2008 - December 2012, N=1827 observations) and one year of data for validation (January - December 2013, N=365 observations). Best-fitting models of ARIMA and linear regression were applied to various 7-day prediction periods and compared using error statistics. The census showed a slightly increasing linear trend. Best fitting models included a non-seasonal model, ARIMA(1,0,0), seasonal ARIMA models, ARIMA(1,0,0)x(1,1,2)7 and ARIMA(2,1,4)x(1,1,2)14, as well as a seasonal linear regression model. Proposed forecasting models resulted on average in 36.49% improvement in forecasting accuracy compared with the fixed average census approach. Time series models provide higher prediction accuracy under different census conditions compared with the fixed average census approach. Presented methodology is easily applicable in clinical practice, can be generalized to other care settings, support short- and long-term census forecasting, and inform staff resource planning.

Comparison and validation of injury risk classifiers for advanced automated crash notification systems.

PubMed

Kusano, Kristofer; Gabler, Hampton C

2014-01-01

The odds of death for a seriously injured crash victim are drastically reduced if he or she received care at a trauma center. Advanced automated crash notification (AACN) algorithms are postcrash safety systems that use data measured by the vehicles during the crash to predict the likelihood of occupants being seriously injured. The accuracy of these models are crucial to the success of an AACN. The objective of this study was to compare the predictive performance of competing injury risk models and algorithms: logistic regression, random forest, AdaBoost, naïve Bayes, support vector machine, and classification k-nearest neighbors. This study compared machine learning algorithms to the widely adopted logistic regression modeling approach. Machine learning algorithms have not been commonly studied in the motor vehicle injury literature. Machine learning algorithms may have higher predictive power than logistic regression, despite the drawback of lacking the ability to perform statistical inference. To evaluate the performance of these algorithms, data on 16,398 vehicles involved in non-rollover collisions were extracted from the NASS-CDS. Vehicles with any occupants having an Injury Severity Score (ISS) of 15 or greater were defined as those requiring victims to be treated at a trauma center. The performance of each model was evaluated using cross-validation. Cross-validation assesses how a model will perform in the future given new data not used for model training. The crash ΔV (change in velocity during the crash), damage side (struck side of the vehicle), seat belt use, vehicle body type, number of events, occupant age, and occupant sex were used as predictors in each model. Logistic regression slightly outperformed the machine learning algorithms based on sensitivity and specificity of the models. Previous studies on AACN risk curves used the same data to train and test the power of the models and as a result had higher sensitivity compared to the cross-validated results from this study. Future studies should account for future data; for example, by using cross-validation or risk presenting optimistic predictions of field performance. Past algorithms have been criticized for relying on age and sex, being difficult to measure by vehicle sensors, and inaccuracies in classifying damage side. The models with accurate damage side and including age/sex did outperform models with less accurate damage side and without age/sex, but the differences were small, suggesting that the success of AACN is not reliant on these predictors.

Functional CAR models for large spatially correlated functional datasets.

PubMed

Zhang, Lin; Baladandayuthapani, Veerabhadran; Zhu, Hongxiao; Baggerly, Keith A; Majewski, Tadeusz; Czerniak, Bogdan A; Morris, Jeffrey S

2016-01-01

We develop a functional conditional autoregressive (CAR) model for spatially correlated data for which functions are collected on areal units of a lattice. Our model performs functional response regression while accounting for spatial correlations with potentially nonseparable and nonstationary covariance structure, in both the space and functional domains. We show theoretically that our construction leads to a CAR model at each functional location, with spatial covariance parameters varying and borrowing strength across the functional domain. Using basis transformation strategies, the nonseparable spatial-functional model is computationally scalable to enormous functional datasets, generalizable to different basis functions, and can be used on functions defined on higher dimensional domains such as images. Through simulation studies, we demonstrate that accounting for the spatial correlation in our modeling leads to improved functional regression performance. Applied to a high-throughput spatially correlated copy number dataset, the model identifies genetic markers not identified by comparable methods that ignore spatial correlations.

On the potential of models for location and scale for genome-wide DNA methylation data

PubMed Central

2014-01-01

Background With the help of epigenome-wide association studies (EWAS), increasing knowledge on the role of epigenetic mechanisms such as DNA methylation in disease processes is obtained. In addition, EWAS aid the understanding of behavioral and environmental effects on DNA methylation. In terms of statistical analysis, specific challenges arise from the characteristics of methylation data. First, methylation β-values represent proportions with skewed and heteroscedastic distributions. Thus, traditional modeling strategies assuming a normally distributed response might not be appropriate. Second, recent evidence suggests that not only mean differences but also variability in site-specific DNA methylation associates with diseases, including cancer. The purpose of this study was to compare different modeling strategies for methylation data in terms of model performance and performance of downstream hypothesis tests. Specifically, we used the generalized additive models for location, scale and shape (GAMLSS) framework to compare beta regression with Gaussian regression on raw, binary logit and arcsine square root transformed methylation data, with and without modeling a covariate effect on the scale parameter. Results Using simulated and real data from a large population-based study and an independent sample of cancer patients and healthy controls, we show that beta regression does not outperform competing strategies in terms of model performance. In addition, Gaussian models for location and scale showed an improved performance as compared to models for location only. The best performance was observed for the Gaussian model on binary logit transformed β-values, referred to as M-values. Our results further suggest that models for location and scale are specifically sensitive towards violations of the distribution assumption and towards outliers in the methylation data. Therefore, a resampling procedure is proposed as a mode of inference and shown to diminish type I error rate in practically relevant settings. We apply the proposed method in an EWAS of BMI and age and reveal strong associations of age with methylation variability that are validated in an independent sample. Conclusions Models for location and scale are promising tools for EWAS that may help to understand the influence of environmental factors and disease-related phenotypes on methylation variability and its role during disease development. PMID:24994026

A statistical learning framework for groundwater nitrate models of the Central Valley, California, USA

USGS Publications Warehouse

Nolan, Bernard T.; Fienen, Michael N.; Lorenz, David L.

2015-01-01

We used a statistical learning framework to evaluate the ability of three machine-learning methods to predict nitrate concentration in shallow groundwater of the Central Valley, California: boosted regression trees (BRT), artificial neural networks (ANN), and Bayesian networks (BN). Machine learning methods can learn complex patterns in the data but because of overfitting may not generalize well to new data. The statistical learning framework involves cross-validation (CV) training and testing data and a separate hold-out data set for model evaluation, with the goal of optimizing predictive performance by controlling for model overfit. The order of prediction performance according to both CV testing R2 and that for the hold-out data set was BRT > BN > ANN. For each method we identified two models based on CV testing results: that with maximum testing R2 and a version with R2 within one standard error of the maximum (the 1SE model). The former yielded CV training R2 values of 0.94–1.0. Cross-validation testing R2 values indicate predictive performance, and these were 0.22–0.39 for the maximum R2 models and 0.19–0.36 for the 1SE models. Evaluation with hold-out data suggested that the 1SE BRT and ANN models predicted better for an independent data set compared with the maximum R2 versions, which is relevant to extrapolation by mapping. Scatterplots of predicted vs. observed hold-out data obtained for final models helped identify prediction bias, which was fairly pronounced for ANN and BN. Lastly, the models were compared with multiple linear regression (MLR) and a previous random forest regression (RFR) model. Whereas BRT results were comparable to RFR, MLR had low hold-out R2 (0.07) and explained less than half the variation in the training data. Spatial patterns of predictions by the final, 1SE BRT model agreed reasonably well with previously observed patterns of nitrate occurrence in groundwater of the Central Valley.

Identifying Autocorrelation Generated by Various Error Processes in Interrupted Time-Series Regression Designs: A Comparison of AR1 and Portmanteau Tests

ERIC Educational Resources Information Center

Huitema, Bradley E.; McKean, Joseph W.

2007-01-01

Regression models used in the analysis of interrupted time-series designs assume statistically independent errors. Four methods of evaluating this assumption are the Durbin-Watson (D-W), Huitema-McKean (H-M), Box-Pierce (B-P), and Ljung-Box (L-B) tests. These tests were compared with respect to Type I error and power under a wide variety of error…

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.

Estimates of nitrate loads and yields from groundwater to streams in the Chesapeake Bay watershed based on land use and geology

USGS Publications Warehouse

Terziotti, Silvia; Capel, Paul D.; Tesoriero, Anthony J.; Hopple, Jessica A.; Kronholm, Scott C.

2018-03-07

The water quality of the Chesapeake Bay may be adversely affected by dissolved nitrate carried in groundwater discharge to streams. To estimate the concentrations, loads, and yields of nitrate from groundwater to streams for the Chesapeake Bay watershed, a regression model was developed based on measured nitrate concentrations from 156 small streams with watersheds less than 500 square miles (mi2 ) at baseflow. The regression model has three predictive variables: geologic unit, percent developed land, and percent agricultural land. Comparisons of estimated and actual values within geologic units were closely matched. The coefficient of determination (R2 ) for the model was 0.6906. The model was used to calculate baseflow nitrate concentrations at over 83,000 National Hydrography Dataset Plus Version 2 catchments and aggregated to 1,966 total 12-digit hydrologic units in the Chesapeake Bay watershed. The modeled output geospatial data layers provided estimated annual loads and yields of nitrate from groundwater into streams. The spatial distribution of annual nitrate yields from groundwater estimated by this method was compared to the total watershed yields of all sources estimated from a Chesapeake Bay SPAtially Referenced Regressions On Watershed attributes (SPARROW) water-quality model. The comparison showed similar spatial patterns. The regression model for groundwater contribution had similar but lower yields, suggesting that groundwater is an important source of nitrogen for streams in the Chesapeake Bay watershed.

The Prediction Properties of Inverse and Reverse Regression for the Simple Linear Calibration Problem

NASA Technical Reports Server (NTRS)

Parker, Peter A.; Geoffrey, Vining G.; Wilson, Sara R.; Szarka, John L., III; Johnson, Nels G.

2010-01-01

The calibration of measurement systems is a fundamental but under-studied problem within industrial statistics. The origins of this problem go back to basic chemical analysis based on NIST standards. In today's world these issues extend to mechanical, electrical, and materials engineering. Often, these new scenarios do not provide "gold standards" such as the standard weights provided by NIST. This paper considers the classic "forward regression followed by inverse regression" approach. In this approach the initial experiment treats the "standards" as the regressor and the observed values as the response to calibrate the instrument. The analyst then must invert the resulting regression model in order to use the instrument to make actual measurements in practice. This paper compares this classical approach to "reverse regression," which treats the standards as the response and the observed measurements as the regressor in the calibration experiment. Such an approach is intuitively appealing because it avoids the need for the inverse regression. However, it also violates some of the basic regression assumptions.

A semi-nonparametric Poisson regression model for analyzing motor vehicle crash data.

PubMed

Ye, Xin; Wang, Ke; Zou, Yajie; Lord, Dominique

2018-01-01

This paper develops a semi-nonparametric Poisson regression model to analyze motor vehicle crash frequency data collected from rural multilane highway segments in California, US. Motor vehicle crash frequency on rural highway is a topic of interest in the area of transportation safety due to higher driving speeds and the resultant severity level. Unlike the traditional Negative Binomial (NB) model, the semi-nonparametric Poisson regression model can accommodate an unobserved heterogeneity following a highly flexible semi-nonparametric (SNP) distribution. Simulation experiments are conducted to demonstrate that the SNP distribution can well mimic a large family of distributions, including normal distributions, log-gamma distributions, bimodal and trimodal distributions. Empirical estimation results show that such flexibility offered by the SNP distribution can greatly improve model precision and the overall goodness-of-fit. The semi-nonparametric distribution can provide a better understanding of crash data structure through its ability to capture potential multimodality in the distribution of unobserved heterogeneity. When estimated coefficients in empirical models are compared, SNP and NB models are found to have a substantially different coefficient for the dummy variable indicating the lane width. The SNP model with better statistical performance suggests that the NB model overestimates the effect of lane width on crash frequency reduction by 83.1%.

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.

Analysis of Blood Transfusion Data Using Bivariate Zero-Inflated Poisson Model: A Bayesian Approach.

PubMed

Mohammadi, Tayeb; Kheiri, Soleiman; Sedehi, Morteza

2016-01-01

Recognizing the factors affecting the number of blood donation and blood deferral has a major impact on blood transfusion. There is a positive correlation between the variables "number of blood donation" and "number of blood deferral": as the number of return for donation increases, so does the number of blood deferral. On the other hand, due to the fact that many donors never return to donate, there is an extra zero frequency for both of the above-mentioned variables. In this study, in order to apply the correlation and to explain the frequency of the excessive zero, the bivariate zero-inflated Poisson regression model was used for joint modeling of the number of blood donation and number of blood deferral. The data was analyzed using the Bayesian approach applying noninformative priors at the presence and absence of covariates. Estimating the parameters of the model, that is, correlation, zero-inflation parameter, and regression coefficients, was done through MCMC simulation. Eventually double-Poisson model, bivariate Poisson model, and bivariate zero-inflated Poisson model were fitted on the data and were compared using the deviance information criteria (DIC). The results showed that the bivariate zero-inflated Poisson regression model fitted the data better than the other models.

Analysis of Blood Transfusion Data Using Bivariate Zero-Inflated Poisson Model: A Bayesian Approach

PubMed Central

Mohammadi, Tayeb; Sedehi, Morteza

2016-01-01

Recognizing the factors affecting the number of blood donation and blood deferral has a major impact on blood transfusion. There is a positive correlation between the variables “number of blood donation” and “number of blood deferral”: as the number of return for donation increases, so does the number of blood deferral. On the other hand, due to the fact that many donors never return to donate, there is an extra zero frequency for both of the above-mentioned variables. In this study, in order to apply the correlation and to explain the frequency of the excessive zero, the bivariate zero-inflated Poisson regression model was used for joint modeling of the number of blood donation and number of blood deferral. The data was analyzed using the Bayesian approach applying noninformative priors at the presence and absence of covariates. Estimating the parameters of the model, that is, correlation, zero-inflation parameter, and regression coefficients, was done through MCMC simulation. Eventually double-Poisson model, bivariate Poisson model, and bivariate zero-inflated Poisson model were fitted on the data and were compared using the deviance information criteria (DIC). The results showed that the bivariate zero-inflated Poisson regression model fitted the data better than the other models. PMID:27703493

Evaluation of the Use of Zero-Augmented Regression Techniques to Model Incidence of Campylobacter Infections in FoodNet.

PubMed

Tremblay, Marlène; Crim, Stacy M; Cole, Dana J; Hoekstra, Robert M; Henao, Olga L; Döpfer, Dörte

2017-10-01

The Foodborne Diseases Active Surveillance Network (FoodNet) is currently using a negative binomial (NB) regression model to estimate temporal changes in the incidence of Campylobacter infection. FoodNet active surveillance in 483 counties collected data on 40,212 Campylobacter cases between years 2004 and 2011. We explored models that disaggregated these data to allow us to account for demographic, geographic, and seasonal factors when examining changes in incidence of Campylobacter infection. We hypothesized that modeling structural zeros and including demographic variables would increase the fit of FoodNet's Campylobacter incidence regression models. Five different models were compared: NB without demographic covariates, NB with demographic covariates, hurdle NB with covariates in the count component only, hurdle NB with covariates in both zero and count components, and zero-inflated NB with covariates in the count component only. Of the models evaluated, the nonzero-augmented NB model with demographic variables provided the best fit. Results suggest that even though zero inflation was not present at this level, individualizing the level of aggregation and using different model structures and predictors per site might be required to correctly distinguish between structural and observational zeros and account for risk factors that vary geographically.

Modeling Personalized Email Prioritization: Classification-based and Regression-based Approaches

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

Yoo S.; Yang, Y.; Carbonell, J.

2011-10-24

Email overload, even after spam filtering, presents a serious productivity challenge for busy professionals and executives. One solution is automated prioritization of incoming emails to ensure the most important are read and processed quickly, while others are processed later as/if time permits in declining priority levels. This paper presents a study of machine learning approaches to email prioritization into discrete levels, comparing ordinal regression versus classier cascades. Given the ordinal nature of discrete email priority levels, SVM ordinal regression would be expected to perform well, but surprisingly a cascade of SVM classifiers significantly outperforms ordinal regression for email prioritization. Inmore » contrast, SVM regression performs well -- better than classifiers -- on selected UCI data sets. This unexpected performance inversion is analyzed and results are presented, providing core functionality for email prioritization systems.« less

Inference for multivariate regression model based on multiply imputed synthetic data generated via posterior predictive sampling

NASA Astrophysics Data System (ADS)

Moura, Ricardo; Sinha, Bimal; Coelho, Carlos A.

2017-06-01

The recent popularity of the use of synthetic data as a Statistical Disclosure Control technique has enabled the development of several methods of generating and analyzing such data, but almost always relying in asymptotic distributions and in consequence being not adequate for small sample datasets. Thus, a likelihood-based exact inference procedure is derived for the matrix of regression coefficients of the multivariate regression model, for multiply imputed synthetic data generated via Posterior Predictive Sampling. Since it is based in exact distributions this procedure may even be used in small sample datasets. Simulation studies compare the results obtained from the proposed exact inferential procedure with the results obtained from an adaptation of Reiters combination rule to multiply imputed synthetic datasets and an application to the 2000 Current Population Survey is discussed.

Analytic model for the long-term evolution of circular Earth satellite orbits including lunar node regression

NASA Astrophysics Data System (ADS)

Zhu, Ting-Lei; Zhao, Chang-Yin; Zhang, Ming-Jiang

2017-04-01

This paper aims to obtain an analytic approximation to the evolution of circular orbits governed by the Earth's J2 and the luni-solar gravitational perturbations. Assuming that the lunar orbital plane coincides with the ecliptic plane, Allan and Cook (Proc. R. Soc. A, Math. Phys. Eng. Sci. 280(1380):97, 1964) derived an analytic solution to the orbital plane evolution of circular orbits. Using their result as an intermediate solution, we establish an approximate analytic model with lunar orbital inclination and its node regression be taken into account. Finally, an approximate analytic expression is derived, which is accurate compared to the numerical results except for the resonant cases when the period of the reference orbit approximately equals the integer multiples (especially 1 or 2 times) of lunar node regression period.

Emergency Department Visit Forecasting and Dynamic Nursing Staff Allocation Using Machine Learning Techniques With Readily Available Open-Source Software.

PubMed

Zlotnik, Alexander; Gallardo-Antolín, Ascensión; Cuchí Alfaro, Miguel; Pérez Pérez, María Carmen; Montero Martínez, Juan Manuel

2015-08-01

Although emergency department visit forecasting can be of use for nurse staff planning, previous research has focused on models that lacked sufficient resolution and realistic error metrics for these predictions to be applied in practice. Using data from a 1100-bed specialized care hospital with 553,000 patients assigned to its healthcare area, forecasts with different prediction horizons, from 2 to 24 weeks ahead, with an 8-hour granularity, using support vector regression, M5P, and stratified average time-series models were generated with an open-source software package. As overstaffing and understaffing errors have different implications, error metrics and potential personnel monetary savings were calculated with a custom validation scheme, which simulated subsequent generation of predictions during a 4-year period. Results were then compared with a generalized estimating equation regression. Support vector regression and M5P models were found to be superior to the stratified average model with a 95% confidence interval. Our findings suggest that medium and severe understaffing situations could be reduced in more than an order of magnitude and average yearly savings of up to €683,500 could be achieved if dynamic nursing staff allocation was performed with support vector regression instead of the static staffing levels currently in use.

Performance and separation occurrence of binary probit regression estimator using maximum likelihood method and Firths approach under different sample size

NASA Astrophysics Data System (ADS)

Lusiana, Evellin Dewi

2017-12-01

The parameters of binary probit regression model are commonly estimated by using Maximum Likelihood Estimation (MLE) method. However, MLE method has limitation if the binary data contains separation. Separation is the condition where there are one or several independent variables that exactly grouped the categories in binary response. It will result the estimators of MLE method become non-convergent, so that they cannot be used in modeling. One of the effort to resolve the separation is using Firths approach instead. This research has two aims. First, to identify the chance of separation occurrence in binary probit regression model between MLE method and Firths approach. Second, to compare the performance of binary probit regression model estimator that obtained by MLE method and Firths approach using RMSE criteria. Those are performed using simulation method and under different sample size. The results showed that the chance of separation occurrence in MLE method for small sample size is higher than Firths approach. On the other hand, for larger sample size, the probability decreased and relatively identic between MLE method and Firths approach. Meanwhile, Firths estimators have smaller RMSE than MLEs especially for smaller sample sizes. But for larger sample sizes, the RMSEs are not much different. It means that Firths estimators outperformed MLE estimator.

Across-Platform Imputation of DNA Methylation Levels Incorporating Nonlocal Information Using Penalized Functional Regression.

PubMed

Zhang, Guosheng; Huang, Kuan-Chieh; Xu, Zheng; Tzeng, Jung-Ying; Conneely, Karen N; Guan, Weihua; Kang, Jian; Li, Yun

2016-05-01

DNA methylation is a key epigenetic mark involved in both normal development and disease progression. Recent advances in high-throughput technologies have enabled genome-wide profiling of DNA methylation. However, DNA methylation profiling often employs different designs and platforms with varying resolution, which hinders joint analysis of methylation data from multiple platforms. In this study, we propose a penalized functional regression model to impute missing methylation data. By incorporating functional predictors, our model utilizes information from nonlocal probes to improve imputation quality. Here, we compared the performance of our functional model to linear regression and the best single probe surrogate in real data and via simulations. Specifically, we applied different imputation approaches to an acute myeloid leukemia dataset consisting of 194 samples and our method showed higher imputation accuracy, manifested, for example, by a 94% relative increase in information content and up to 86% more CpG sites passing post-imputation filtering. Our simulated association study further demonstrated that our method substantially improves the statistical power to identify trait-associated methylation loci. These findings indicate that the penalized functional regression model is a convenient and valuable imputation tool for methylation data, and it can boost statistical power in downstream epigenome-wide association study (EWAS). © 2016 WILEY PERIODICALS, INC.

Integrating Address Geocoding, Land Use Regression, and Spatiotemporal Geostatistical Estimation for Groundwater Tetrachloroethylene

PubMed Central

Messier, Kyle P.; Akita, Yasuyuki; Serre, Marc L.

2012-01-01

Geographic Information Systems (GIS) based techniques are cost-effective and efficient methods used by state agencies and epidemiology researchers for estimating concentration and exposure. However, budget limitations have made statewide assessments of contamination difficult, especially in groundwater media. Many studies have implemented address geocoding, land use regression, and geostatistics independently, but this is the first to examine the benefits of integrating these GIS techniques to address the need of statewide exposure assessments. A novel framework for concentration exposure is introduced that integrates address geocoding, land use regression (LUR), below detect data modeling, and Bayesian Maximum Entropy (BME). A LUR model was developed for Tetrachloroethylene that accounts for point sources and flow direction. We then integrate the LUR model into the BME method as a mean trend while also modeling below detects data as a truncated Gaussian probability distribution function. We increase available PCE data 4.7 times from previously available databases through multistage geocoding. The LUR model shows significant influence of dry cleaners at short ranges. The integration of the LUR model as mean trend in BME results in a 7.5% decrease in cross validation mean square error compared to BME with a constant mean trend. PMID:22264162

Integrating address geocoding, land use regression, and spatiotemporal geostatistical estimation for groundwater tetrachloroethylene.

PubMed

Messier, Kyle P; Akita, Yasuyuki; Serre, Marc L

2012-03-06

Geographic information systems (GIS) based techniques are cost-effective and efficient methods used by state agencies and epidemiology researchers for estimating concentration and exposure. However, budget limitations have made statewide assessments of contamination difficult, especially in groundwater media. Many studies have implemented address geocoding, land use regression, and geostatistics independently, but this is the first to examine the benefits of integrating these GIS techniques to address the need of statewide exposure assessments. A novel framework for concentration exposure is introduced that integrates address geocoding, land use regression (LUR), below detect data modeling, and Bayesian Maximum Entropy (BME). A LUR model was developed for tetrachloroethylene that accounts for point sources and flow direction. We then integrate the LUR model into the BME method as a mean trend while also modeling below detects data as a truncated Gaussian probability distribution function. We increase available PCE data 4.7 times from previously available databases through multistage geocoding. The LUR model shows significant influence of dry cleaners at short ranges. The integration of the LUR model as mean trend in BME results in a 7.5% decrease in cross validation mean square error compared to BME with a constant mean trend.

A Comprehensive review of group level model performance in the presence of heteroscedasticity: Can a single model control Type I errors in the presence of outliers?

PubMed Central

Mumford, Jeanette A.

2017-01-01

Even after thorough preprocessing and a careful time series analysis of functional magnetic resonance imaging (fMRI) data, artifact and other issues can lead to violations of the assumption that the variance is constant across subjects in the group level model. This is especially concerning when modeling a continuous covariate at the group level, as the slope is easily biased by outliers. Various models have been proposed to deal with outliers including models that use the first level variance or that use the group level residual magnitude to differentially weight subjects. The most typically used robust regression, implementing a robust estimator of the regression slope, has been previously studied in the context of fMRI studies and was found to perform well in some scenarios, but a loss of Type I error control can occur for some outlier settings. A second type of robust regression using a heteroscedastic autocorrelation consistent (HAC) estimator, which produces robust slope and variance estimates has been shown to perform well, with better Type I error control, but with large sample sizes (500–1000 subjects). The Type I error control with smaller sample sizes has not been studied in this model and has not been compared to other modeling approaches that handle outliers such as FSL’s Flame 1 and FSL’s outlier de-weighting. Focusing on group level inference with a continuous covariate over a range of sample sizes and degree of heteroscedasticity, which can be driven either by the within- or between-subject variability, both styles of robust regression are compared to ordinary least squares (OLS), FSL’s Flame 1, Flame 1 with outlier de-weighting algorithm and Kendall’s Tau. Additionally, subject omission using the Cook’s Distance measure with OLS and nonparametric inference with the OLS statistic are studied. Pros and cons of these models as well as general strategies for detecting outliers in data and taking precaution to avoid inflated Type I error rates are discussed. PMID:28030782

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.

Predicting spatio-temporal failure in large scale observational and micro scale experimental systems

NASA Astrophysics Data System (ADS)

de las Heras, Alejandro; Hu, Yong

2006-10-01

Forecasting has become an essential part of modern thought, but the practical limitations still are manifold. We addressed future rates of change by comparing models that take into account time, and models that focus more on space. Cox regression confirmed that linear change can be safely assumed in the short-term. Spatially explicit Poisson regression, provided a ceiling value for the number of deforestation spots. With several observed and estimated rates, it was decided to forecast using the more robust assumptions. A Markov-chain cellular automaton thus projected 5-year deforestation in the Amazonian Arc of Deforestation, showing that even a stable rate of change would largely deplete the forest area. More generally, resolution and implementation of the existing models could explain many of the modelling difficulties still affecting forecasting.

Improving the Spatial Prediction of Soil Organic Carbon Stocks in a Complex Tropical Mountain Landscape by Methodological Specifications in Machine Learning Approaches.

PubMed

Ließ, Mareike; Schmidt, Johannes; Glaser, Bruno

2016-01-01

Tropical forests are significant carbon sinks and their soils' carbon storage potential is immense. However, little is known about the soil organic carbon (SOC) stocks of tropical mountain areas whose complex soil-landscape and difficult accessibility pose a challenge to spatial analysis. The choice of methodology for spatial prediction is of high importance to improve the expected poor model results in case of low predictor-response correlations. Four aspects were considered to improve model performance in predicting SOC stocks of the organic layer of a tropical mountain forest landscape: Different spatial predictor settings, predictor selection strategies, various machine learning algorithms and model tuning. Five machine learning algorithms: random forests, artificial neural networks, multivariate adaptive regression splines, boosted regression trees and support vector machines were trained and tuned to predict SOC stocks from predictors derived from a digital elevation model and satellite image. Topographical predictors were calculated with a GIS search radius of 45 to 615 m. Finally, three predictor selection strategies were applied to the total set of 236 predictors. All machine learning algorithms-including the model tuning and predictor selection-were compared via five repetitions of a tenfold cross-validation. The boosted regression tree algorithm resulted in the overall best model. SOC stocks ranged between 0.2 to 17.7 kg m-2, displaying a huge variability with diffuse insolation and curvatures of different scale guiding the spatial pattern. Predictor selection and model tuning improved the models' predictive performance in all five machine learning algorithms. The rather low number of selected predictors favours forward compared to backward selection procedures. Choosing predictors due to their indiviual performance was vanquished by the two procedures which accounted for predictor interaction.

Simulation on Poisson and negative binomial models of count road accident modeling

NASA Astrophysics Data System (ADS)

Sapuan, M. S.; Razali, A. M.; Zamzuri, Z. H.; Ibrahim, K.

2016-11-01

Accident count data have often been shown to have overdispersion. On the other hand, the data might contain zero count (excess zeros). The simulation study was conducted to create a scenarios which an accident happen in T-junction with the assumption the dependent variables of generated data follows certain distribution namely Poisson and negative binomial distribution with different sample size of n=30 to n=500. The study objective was accomplished by fitting Poisson regression, negative binomial regression and Hurdle negative binomial model to the simulated data. The model validation was compared and the simulation result shows for each different sample size, not all model fit the data nicely even though the data generated from its own distribution especially when the sample size is larger. Furthermore, the larger sample size indicates that more zeros accident count in the dataset.

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.

Regression models to predict hip joint centers in pathological hip population.

PubMed

Mantovani, Giulia; Ng, K C Geoffrey; Lamontagne, Mario

2016-02-01

The purpose was to investigate the validity of Harrington's and Davis's hip joint center (HJC) regression equations on a population affected by a hip deformity, (i.e., femoroacetabular impingement). Sixty-seven participants (21 healthy controls, 46 with a cam-type deformity) underwent pelvic CT imaging. Relevant bony landmarks and geometric HJCs were digitized from the images, and skin thickness was measured for the anterior and posterior superior iliac spines. Non-parametric statistical and Bland-Altman tests analyzed differences between the predicted HJC (from regression equations) and the actual HJC (from CT images). The error from Davis's model (25.0 ± 6.7 mm) was larger than Harrington's (12.3 ± 5.9 mm, p<0.001). There were no differences between groups, thus, studies on femoroacetabular impingement can implement conventional regression models. Measured skin thickness was 9.7 ± 7.0mm and 19.6 ± 10.9 mm for the anterior and posterior bony landmarks, respectively, and correlated with body mass index. Skin thickness estimates can be considered to reduce the systematic error introduced by surface markers. New adult-specific regression equations were developed from the CT dataset, with the hypothesis that they could provide better estimates when tuned to a larger adult-specific dataset. The linear models were validated on external datasets and using leave-one-out cross-validation techniques; Prediction errors were comparable to those of Harrington's model, despite the adult-specific population and the larger sample size, thus, prediction accuracy obtained from these parameters could not be improved. Copyright © 2015 Elsevier B.V. All rights reserved.

Poisson Mixture Regression Models for Heart Disease Prediction.

PubMed

Mufudza, Chipo; Erol, Hamza

2016-01-01

Early heart disease control can be achieved by high disease prediction and diagnosis efficiency. This paper focuses on the use of model based clustering techniques to predict and diagnose heart disease via Poisson mixture regression models. Analysis and application of Poisson mixture regression models is here addressed under two different classes: standard and concomitant variable mixture regression models. Results show that a two-component concomitant variable Poisson mixture regression model predicts heart disease better than both the standard Poisson mixture regression model and the ordinary general linear Poisson regression model due to its low Bayesian Information Criteria value. Furthermore, a Zero Inflated Poisson Mixture Regression model turned out to be the best model for heart prediction over all models as it both clusters individuals into high or low risk category and predicts rate to heart disease componentwise given clusters available. It is deduced that heart disease prediction can be effectively done by identifying the major risks componentwise using Poisson mixture regression model.

Poisson Mixture Regression Models for Heart Disease Prediction

PubMed Central

Erol, Hamza

2016-01-01

Early heart disease control can be achieved by high disease prediction and diagnosis efficiency. This paper focuses on the use of model based clustering techniques to predict and diagnose heart disease via Poisson mixture regression models. Analysis and application of Poisson mixture regression models is here addressed under two different classes: standard and concomitant variable mixture regression models. Results show that a two-component concomitant variable Poisson mixture regression model predicts heart disease better than both the standard Poisson mixture regression model and the ordinary general linear Poisson regression model due to its low Bayesian Information Criteria value. Furthermore, a Zero Inflated Poisson Mixture Regression model turned out to be the best model for heart prediction over all models as it both clusters individuals into high or low risk category and predicts rate to heart disease componentwise given clusters available. It is deduced that heart disease prediction can be effectively done by identifying the major risks componentwise using Poisson mixture regression model. PMID:27999611

The allometry of coarse root biomass: log-transformed linear regression or nonlinear regression?

PubMed

Lai, Jiangshan; Yang, Bo; Lin, Dunmei; Kerkhoff, Andrew J; Ma, Keping

2013-01-01

Precise estimation of root biomass is important for understanding carbon stocks and dynamics in forests. Traditionally, biomass estimates are based on allometric scaling relationships between stem diameter and coarse root biomass calculated using linear regression (LR) on log-transformed data. Recently, it has been suggested that nonlinear regression (NLR) is a preferable fitting method for scaling relationships. But while this claim has been contested on both theoretical and empirical grounds, and statistical methods have been developed to aid in choosing between the two methods in particular cases, few studies have examined the ramifications of erroneously applying NLR. Here, we use direct measurements of 159 trees belonging to three locally dominant species in east China to compare the LR and NLR models of diameter-root biomass allometry. We then contrast model predictions by estimating stand coarse root biomass based on census data from the nearby 24-ha Gutianshan forest plot and by testing the ability of the models to predict known root biomass values measured on multiple tropical species at the Pasoh Forest Reserve in Malaysia. Based on likelihood estimates for model error distributions, as well as the accuracy of extrapolative predictions, we find that LR on log-transformed data is superior to NLR for fitting diameter-root biomass scaling models. More importantly, inappropriately using NLR leads to grossly inaccurate stand biomass estimates, especially for stands dominated by smaller trees.

Bias correction by use of errors-in-variables regression models in studies with K-X-ray fluorescence bone lead measurements.

PubMed

Lamadrid-Figueroa, Héctor; Téllez-Rojo, Martha M; Angeles, Gustavo; Hernández-Ávila, Mauricio; Hu, Howard

2011-01-01

In-vivo measurement of bone lead by means of K-X-ray fluorescence (KXRF) is the preferred biological marker of chronic exposure to lead. Unfortunately, considerable measurement error associated with KXRF estimations can introduce bias in estimates of the effect of bone lead when this variable is included as the exposure in a regression model. Estimates of uncertainty reported by the KXRF instrument reflect the variance of the measurement error and, although they can be used to correct the measurement error bias, they are seldom used in epidemiological statistical analyzes. Errors-in-variables regression (EIV) allows for correction of bias caused by measurement error in predictor variables, based on the knowledge of the reliability of such variables. The authors propose a way to obtain reliability coefficients for bone lead measurements from uncertainty data reported by the KXRF instrument and compare, by the use of Monte Carlo simulations, results obtained using EIV regression models vs. those obtained by the standard procedures. Results of the simulations show that Ordinary Least Square (OLS) regression models provide severely biased estimates of effect, and that EIV provides nearly unbiased estimates. Although EIV effect estimates are more imprecise, their mean squared error is much smaller than that of OLS estimates. In conclusion, EIV is a better alternative than OLS to estimate the effect of bone lead when measured by KXRF. Copyright © 2010 Elsevier Inc. All rights reserved.

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.

Selecting a Variable for Predicting the Diagnosis of PTB Patients From Comparison of Chest X-ray Images

NASA Astrophysics Data System (ADS)

Mohd. Rijal, Omar; Mohd. Noor, Norliza; Teng, Shee Lee

A statistical method of comparing two digital chest radiographs for Pulmonary Tuberculosis (PTB) patients has been proposed. After applying appropriate image registration procedures, a selected subset of each image is converted to an image histogram (or box plot). Comparing two chest X-ray images is equivalent to the direct comparison of the two corresponding histograms. From each histogram, eleven percentiles (of image intensity) are calculated. The number of percentiles that shift to the left (NLSP) when second image is compared to the first has been shown to be an indicator of patients` progress. In this study, the values of NLSP is to be compared with the actual diagnosis (Y) of several medical practitioners. A logistic regression model is used to study the relationship between NLSP and Y. This study showed that NLSP may be used as an alternative or second opinion for Y. The proposed regression model also show that important explanatory variables such as outcomes of sputum test (Z) and degree of image registration (W) may be omitted when estimating Y-values.

Membrane Introduction Mass Spectrometry Combined with an Orthogonal Partial-Least Squares Calibration Model for Mixture Analysis.

PubMed

Li, Min; Zhang, Lu; Yao, Xiaolong; Jiang, Xingyu

2017-01-01

The emerging membrane introduction mass spectrometry technique has been successfully used to detect benzene, toluene, ethyl benzene and xylene (BTEX), while overlapped spectra have unfortunately hindered its further application to the analysis of mixtures. Multivariate calibration, an efficient method to analyze mixtures, has been widely applied. In this paper, we compared univariate and multivariate analyses for quantification of the individual components of mixture samples. The results showed that the univariate analysis creates poor models with regression coefficients of 0.912, 0.867, 0.440 and 0.351 for BTEX, respectively. For multivariate analysis, a comparison to the partial-least squares (PLS) model shows that the orthogonal partial-least squares (OPLS) regression exhibits an optimal performance with regression coefficients of 0.995, 0.999, 0.980 and 0.976, favorable calibration parameters (RMSEC and RMSECV) and a favorable validation parameter (RMSEP). Furthermore, the OPLS exhibits a good recovery of 73.86 - 122.20% and relative standard deviation (RSD) of the repeatability of 1.14 - 4.87%. Thus, MIMS coupled with the OPLS regression provides an optimal approach for a quantitative BTEX mixture analysis in monitoring and predicting water pollution.

Evaluating lake phytoplanton response to human disturbance and climate change using satellite imagery

NASA Astrophysics Data System (ADS)

Novitski, Linda Nicole

Accurate and cost-effective assessment of water quality is necessary for proper management and restoration of inland water bodies susceptible to algal bloom conditions. Landsat and MODIS satellite images were used to create chlorophyll and Secchi depth predictive models for algal assessment of Great Lakes and other lakes of the United States. Boosted regression tree (BRT) models using satellite imagery are both easy to use and can have high predictive performance. BRT models inferred chlorophyll and Secchi depth more accurately than linear regression models for all study locations. Inferred chlorophyll of inner Saginaw Bay was subsequently used in ecological models to help understand the ecological drivers of algal blooms in this ecosystem. For small lakes (non-Great Lakes), the best national Landsat model for ln-transformed chlorophyll was the BRT model and had a cross-validation R 2 of 0.44 and a 0.76 ln-transformed mug/L RMSE. The best national Landsat model for Secchi depth was also a BRT model that had an adjusted R 2 of 0.52 and a 0.80 m RMSE. We assessed the applicability of the national chlorophyll model for ecological analysis by comparing the total phosphorus- chlorophyll relationship with chlorophyll determined from sampling or remote sensing, which showed the total phosphorus- chlorophyll relationship had an adjusted R2 = 0.58 and 1.02 ln-transformed microg/L RMSE with sampled chlorophyll versus an adjusted R2 = 0.56 and 1.04 ln-transformed mug/L RMSE with chlorophyll determined by the boosted regression tree remote sensing model. For Great Lakes models, the MODIS BRT model predicted chlorophyll most accurately of the three BRT models and compared well to other models in the literature. BRT models for Landsat ETM+ and TM more accurately predicted chlorophyll than the MSS model and all Landsat models had favorable results when compared to the literature. BRT chlorophyll predictive models are useful in helping to understand historical, long-term chlorophyll trends and to inform us of how climate change may alter ecosystems in the future. In inner Saginaw Bay, annual average and upper quartile Landsat-derived chlorophyll decreased from 7.44 to 6.62 and 8.38 to 7.38 mug/L between 1973-1982, and annual upper quartile of 8-day phosphorus loads increased from 5.29 to 6.79 kg between 1973-2012. Simple linear and multiple regression models and Wilcoxon rank test results for MODIS and Landsat-derived chlorophyll indicate that distance from the Saginaw River mouth influences chlorophyll concentration in Saginaw Bay; Landsat-derived surface water temperature and phosphorus loads to a lesser extent. Mixed-effect models for MODIS and Landsat-derived chlorophyll were related to chlorophyll better than simple linear or multiple regressions, with random effects of pixel and sample date contributing substantially to predictive power (NSE=0.35-70), though phosphorus loads, distance to Saginaw River mouth, and water were significant fixed effects in most models. Water quality changes in Saginaw Bay between 1972-2012 were influenced by phosphorus loading and distance to the Saginaw River's mouth. Landsat and MODIS imagery are complementary platforms because of the long history of Landsat operation and the finer spectral resolution and image frequency of MODIS. Remote sensing water quality assessment tools can be valuable for limnological study, ecological assessment, and water resource management.

Fast function-on-scalar regression with penalized basis expansions.

PubMed

Reiss, Philip T; Huang, Lei; Mennes, Maarten

2010-01-01

Regression models for functional responses and scalar predictors are often fitted by means of basis functions, with quadratic roughness penalties applied to avoid overfitting. The fitting approach described by Ramsay and Silverman in the 1990 s amounts to a penalized ordinary least squares (P-OLS) estimator of the coefficient functions. We recast this estimator as a generalized ridge regression estimator, and present a penalized generalized least squares (P-GLS) alternative. We describe algorithms by which both estimators can be implemented, with automatic selection of optimal smoothing parameters, in a more computationally efficient manner than has heretofore been available. We discuss pointwise confidence intervals for the coefficient functions, simultaneous inference by permutation tests, and model selection, including a novel notion of pointwise model selection. P-OLS and P-GLS are compared in a simulation study. Our methods are illustrated with an analysis of age effects in a functional magnetic resonance imaging data set, as well as a reanalysis of a now-classic Canadian weather data set. An R package implementing the methods is publicly available.

Prediction of sweetness and amino acid content in soybean crops from hyperspectral imagery

NASA Astrophysics Data System (ADS)

Monteiro, Sildomar Takahashi; Minekawa, Yohei; Kosugi, Yukio; Akazawa, Tsuneya; Oda, Kunio

Hyperspectral image data provides a powerful tool for non-destructive crop analysis. This paper investigates a hyperspectral image data-processing method to predict the sweetness and amino acid content of soybean crops. Regression models based on artificial neural networks were developed in order to calculate the level of sucrose, glucose, fructose, and nitrogen concentrations, which can be related to the sweetness and amino acid content of vegetables. A performance analysis was conducted comparing regression models obtained using different preprocessing methods, namely, raw reflectance, second derivative, and principal components analysis. This method is demonstrated using high-resolution hyperspectral data of wavelengths ranging from the visible to the near infrared acquired from an experimental field of green vegetable soybeans. The best predictions were achieved using a nonlinear regression model of the second derivative transformed dataset. Glucose could be predicted with greater accuracy, followed by sucrose, fructose and nitrogen. The proposed method provides the possibility to provide relatively accurate maps predicting the chemical content of soybean crop fields.

The Use of Linear Programming for Prediction.

ERIC Educational Resources Information Center

Schnittjer, Carl J.

The purpose of the study was to develop a linear programming model to be used for prediction, test the accuracy of the predictions, and compare the accuracy with that produced by curvilinear multiple regression analysis. (Author)

Predictors of the number of under-five malnourished children in Bangladesh: application of the generalized poisson regression model

PubMed Central

2013-01-01

Background Malnutrition is one of the principal causes of child mortality in developing countries including Bangladesh. According to our knowledge, most of the available studies, that addressed the issue of malnutrition among under-five children, considered the categorical (dichotomous/polychotomous) outcome variables and applied logistic regression (binary/multinomial) to find their predictors. In this study malnutrition variable (i.e. outcome) is defined as the number of under-five malnourished children in a family, which is a non-negative count variable. The purposes of the study are (i) to demonstrate the applicability of the generalized Poisson regression (GPR) model as an alternative of other statistical methods and (ii) to find some predictors of this outcome variable. Methods The data is extracted from the Bangladesh Demographic and Health Survey (BDHS) 2007. Briefly, this survey employs a nationally representative sample which is based on a two-stage stratified sample of households. A total of 4,460 under-five children is analysed using various statistical techniques namely Chi-square test and GPR model. Results The GPR model (as compared to the standard Poisson regression and negative Binomial regression) is found to be justified to study the above-mentioned outcome variable because of its under-dispersion (variance < mean) property. Our study also identify several significant predictors of the outcome variable namely mother’s education, father’s education, wealth index, sanitation status, source of drinking water, and total number of children ever born to a woman. Conclusions Consistencies of our findings in light of many other studies suggest that the GPR model is an ideal alternative of other statistical models to analyse the number of under-five malnourished children in a family. Strategies based on significant predictors may improve the nutritional status of children in Bangladesh. PMID:23297699

Tutorial on Biostatistics: Linear Regression Analysis of Continuous Correlated Eye Data

PubMed Central

Ying, Gui-shuang; Maguire, Maureen G; Glynn, Robert; Rosner, Bernard

2017-01-01

Purpose To describe and demonstrate appropriate linear regression methods for analyzing correlated continuous eye data. Methods We describe several approaches to regression analysis involving both eyes, including mixed effects and marginal models under various covariance structures to account for inter-eye correlation. We demonstrate, with SAS statistical software, applications in a study comparing baseline refractive error between one eye with choroidal neovascularization (CNV) and the unaffected fellow eye, and in a study determining factors associated with visual field data in the elderly. Results When refractive error from both eyes were analyzed with standard linear regression without accounting for inter-eye correlation (adjusting for demographic and ocular covariates), the difference between eyes with CNV and fellow eyes was 0.15 diopters (D; 95% confidence interval, CI −0.03 to 0.32D, P=0.10). Using a mixed effects model or a marginal model, the estimated difference was the same but with narrower 95% CI (0.01 to 0.28D, P=0.03). Standard regression for visual field data from both eyes provided biased estimates of standard error (generally underestimated) and smaller P-values, while analysis of the worse eye provided larger P-values than mixed effects models and marginal models. Conclusion In research involving both eyes, ignoring inter-eye correlation can lead to invalid inferences. Analysis using only right or left eyes is valid, but decreases power. Worse-eye analysis can provide less power and biased estimates of effect. Mixed effects or marginal models using the eye as the unit of analysis should be used to appropriately account for inter-eye correlation and maximize power and precision. PMID:28102741

New regression model for predicting hand-arm vibration (HAV) of Malaysian Army (MA) three-tonne truck steering wheels.

PubMed

Aziz, Shamsul Akmar Ab; Nuawi, Mohd Zaki; Nor, Mohd Jailani Mohd

2015-01-01

The objective of this study was to present a new method for determination of hand-arm vibration (HAV) in Malaysian Army (MA) three-tonne truck steering wheels based on changes in vehicle speed using regression model and the statistical analysis method known as Integrated Kurtosis-Based Algorithm for Z-Notch Filter Technique Vibro (I-kaz Vibro). The test was conducted for two different road conditions, tarmac and dirt roads. HAV exposure was measured using a Brüel & Kjær Type 3649 vibration analyzer, which is capable of recording HAV exposures from steering wheels. The data was analyzed using I-kaz Vibro to determine the HAV values in relation to varying speeds of a truck and to determine the degree of data scattering for HAV data signals. Based on the results obtained, HAV experienced by drivers can be determined using the daily vibration exposure A(8), I-kaz Vibro coefficient (Ƶ(v)(∞)), and the I-kaz Vibro display. The I-kaz Vibro displays also showed greater scatterings, indicating that the values of Ƶ(v)(∞) and A(8) were increasing. Prediction of HAV exposure was done using the developed regression model and graphical representations of Ƶ(v)(∞). The results of the regression model showed that Ƶ(v)(∞) increased when the vehicle speed and HAV exposure increased. For model validation, predicted and measured noise exposures were compared, and high coefficient of correlation (R(2)) values were obtained, indicating that good agreement was obtained between them. By using the developed regression model, we can easily predict HAV exposure from steering wheels for HAV exposure monitoring.

Statistical procedures for evaluating daily and monthly hydrologic model predictions

USGS Publications Warehouse

Coffey, M.E.; Workman, S.R.; Taraba, J.L.; Fogle, A.W.

2004-01-01

The overall study objective was to evaluate the applicability of different qualitative and quantitative methods for comparing daily and monthly SWAT computer model hydrologic streamflow predictions to observed data, and to recommend statistical methods for use in future model evaluations. Statistical methods were tested using daily streamflows and monthly equivalent runoff depths. The statistical techniques included linear regression, Nash-Sutcliffe efficiency, nonparametric tests, t-test, objective functions, autocorrelation, and cross-correlation. None of the methods specifically applied to the non-normal distribution and dependence between data points for the daily predicted and observed data. Of the tested methods, median objective functions, sign test, autocorrelation, and cross-correlation were most applicable for the daily data. The robust coefficient of determination (CD*) and robust modeling efficiency (EF*) objective functions were the preferred methods for daily model results due to the ease of comparing these values with a fixed ideal reference value of one. Predicted and observed monthly totals were more normally distributed, and there was less dependence between individual monthly totals than was observed for the corresponding predicted and observed daily values. More statistical methods were available for comparing SWAT model-predicted and observed monthly totals. The 1995 monthly SWAT model predictions and observed data had a regression Rr2 of 0.70, a Nash-Sutcliffe efficiency of 0.41, and the t-test failed to reject the equal data means hypothesis. The Nash-Sutcliffe coefficient and the R r2 coefficient were the preferred methods for monthly results due to the ability to compare these coefficients to a set ideal value of one.

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

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.

Connectome-based predictive modeling of attention: Comparing different functional connectivity features and prediction methods across datasets.

PubMed

Yoo, Kwangsun; Rosenberg, Monica D; Hsu, Wei-Ting; Zhang, Sheng; Li, Chiang-Shan R; Scheinost, Dustin; Constable, R Todd; Chun, Marvin M

2018-02-15

Connectome-based predictive modeling (CPM; Finn et al., 2015; Shen et al., 2017) was recently developed to predict individual differences in traits and behaviors, including fluid intelligence (Finn et al., 2015) and sustained attention (Rosenberg et al., 2016a), from functional brain connectivity (FC) measured with fMRI. Here, using the CPM framework, we compared the predictive power of three different measures of FC (Pearson's correlation, accordance, and discordance) and two different prediction algorithms (linear and partial least square [PLS] regression) for attention function. Accordance and discordance are recently proposed FC measures that respectively track in-phase synchronization and out-of-phase anti-correlation (Meskaldji et al., 2015). We defined connectome-based models using task-based or resting-state FC data, and tested the effects of (1) functional connectivity measure and (2) feature-selection/prediction algorithm on individualized attention predictions. Models were internally validated in a training dataset using leave-one-subject-out cross-validation, and externally validated with three independent datasets. The training dataset included fMRI data collected while participants performed a sustained attention task and rested (N = 25; Rosenberg et al., 2016a). The validation datasets included: 1) data collected during performance of a stop-signal task and at rest (N = 83, including 19 participants who were administered methylphenidate prior to scanning; Farr et al., 2014a; Rosenberg et al., 2016b), 2) data collected during Attention Network Task performance and rest (N = 41, Rosenberg et al., in press), and 3) resting-state data and ADHD symptom severity from the ADHD-200 Consortium (N = 113; Rosenberg et al., 2016a). Models defined using all combinations of functional connectivity measure (Pearson's correlation, accordance, and discordance) and prediction algorithm (linear and PLS regression) predicted attentional abilities, with correlations between predicted and observed measures of attention as high as 0.9 for internal validation, and 0.6 for external validation (all p's < 0.05). Models trained on task data outperformed models trained on rest data. Pearson's correlation and accordance features generally showed a small numerical advantage over discordance features, while PLS regression models were usually better than linear regression models. Overall, in addition to correlation features combined with linear models (Rosenberg et al., 2016a), it is useful to consider accordance features and PLS regression for CPM. Copyright © 2017 Elsevier Inc. All rights reserved.

A retrospective analysis to identify the factors affecting infection in patients undergoing chemotherapy.

PubMed

Park, Ji Hyun; Kim, Hyeon-Young; Lee, Hanna; Yun, Eun Kyoung

2015-12-01

This study compares the performance of the logistic regression and decision tree analysis methods for assessing the risk factors for infection in cancer patients undergoing chemotherapy. The subjects were 732 cancer patients who were receiving chemotherapy at K university hospital in Seoul, Korea. The data were collected between March 2011 and February 2013 and were processed for descriptive analysis, logistic regression and decision tree analysis using the IBM SPSS Statistics 19 and Modeler 15.1 programs. The most common risk factors for infection in cancer patients receiving chemotherapy were identified as alkylating agents, vinca alkaloid and underlying diabetes mellitus. The logistic regression explained 66.7% of the variation in the data in terms of sensitivity and 88.9% in terms of specificity. The decision tree analysis accounted for 55.0% of the variation in the data in terms of sensitivity and 89.0% in terms of specificity. As for the overall classification accuracy, the logistic regression explained 88.0% and the decision tree analysis explained 87.2%. The logistic regression analysis showed a higher degree of sensitivity and classification accuracy. Therefore, logistic regression analysis is concluded to be the more effective and useful method for establishing an infection prediction model for patients undergoing chemotherapy. Copyright © 2015 Elsevier Ltd. All rights reserved.

A hybrid neural network model for noisy data regression.

PubMed

Lee, Eric W M; Lim, Chee Peng; Yuen, Richard K K; Lo, S M

2004-04-01

A hybrid neural network model, based on the fusion of fuzzy adaptive resonance theory (FA ART) and the general regression neural network (GRNN), is proposed in this paper. Both FA and the GRNN are incremental learning systems and are very fast in network training. The proposed hybrid model, denoted as GRNNFA, is able to retain these advantages and, at the same time, to reduce the computational requirements in calculating and storing information of the kernels. A clustering version of the GRNN is designed with data compression by FA for noise removal. An adaptive gradient-based kernel width optimization algorithm has also been devised. Convergence of the gradient descent algorithm can be accelerated by the geometric incremental growth of the updating factor. A series of experiments with four benchmark datasets have been conducted to assess and compare effectiveness of GRNNFA with other approaches. The GRNNFA model is also employed in a novel application task for predicting the evacuation time of patrons at typical karaoke centers in Hong Kong in the event of fire. The results positively demonstrate the applicability of GRNNFA in noisy data regression problems.

Predictive landslide susceptibility mapping using spatial information in the Pechabun area of Thailand

NASA Astrophysics Data System (ADS)

Oh, Hyun-Joo; Lee, Saro; Chotikasathien, Wisut; Kim, Chang Hwan; Kwon, Ju Hyoung

2009-04-01

For predictive landslide susceptibility mapping, this study applied and verified probability model, the frequency ratio and statistical model, logistic regression at Pechabun, Thailand, using a geographic information system (GIS) and remote sensing. Landslide locations were identified in the study area from interpretation of aerial photographs and field surveys, and maps of the topography, geology and land cover were constructed to spatial database. The factors that influence landslide occurrence, such as slope gradient, slope aspect and curvature of topography and distance from drainage were calculated from the topographic database. Lithology and distance from fault were extracted and calculated from the geology database. Land cover was classified from Landsat TM satellite image. The frequency ratio and logistic regression coefficient were overlaid for landslide susceptibility mapping as each factor’s ratings. Then the landslide susceptibility map was verified and compared using the existing landslide location. As the verification results, the frequency ratio model showed 76.39% and logistic regression model showed 70.42% in prediction accuracy. The method can be used to reduce hazards associated with landslides and to plan land cover.

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.

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

Bias and uncertainty of δ13CO2 isotopic mixing models

Treesearch

Zachary E. Kayler; Lisa Ganio; Mark Hauck; Thomas G. Pypker; Elizabeth W. Sulzman; Alan C. Mix; Barbara J. Bond

2009-01-01

The goal of this study was to evaluate how factorial combinations of two mixing models and two regression approaches (Keeling-OLS, Miller—Tans-OLS, Keeling-GMR, Miller—Tans-GMR) compare in small [CO2] range versus large[CO2] range regimes, with different combinations of...

Forecasting Enrollments with Fuzzy Time Series.

ERIC Educational Resources Information Center

Song, Qiang; Chissom, Brad S.

The concept of fuzzy time series is introduced and used to forecast the enrollment of a university. Fuzzy time series, an aspect of fuzzy set theory, forecasts enrollment using a first-order time-invariant model. To evaluate the model, the conventional linear regression technique is applied and the predicted values obtained are compared to the…

Weakly Informative Prior for Point Estimation of Covariance Matrices in Hierarchical Models

ERIC Educational Resources Information Center

Chung, Yeojin; Gelman, Andrew; Rabe-Hesketh, Sophia; Liu, Jingchen; Dorie, Vincent

2015-01-01

When fitting hierarchical regression models, maximum likelihood (ML) estimation has computational (and, for some users, philosophical) advantages compared to full Bayesian inference, but when the number of groups is small, estimates of the covariance matrix (S) of group-level varying coefficients are often degenerate. One can do better, even from…

Carbon dioxide stripping in aquaculture -- part III: model verification

USGS Publications Warehouse

Colt, John; Watten, Barnaby; Pfeiffer, Tim

2012-01-01

Based on conventional mass transfer models developed for oxygen, the use of the non-linear ASCE method, 2-point method, and one parameter linear-regression method were evaluated for carbon dioxide stripping data. For values of KLaCO2 < approximately 1.5/h, the 2-point or ASCE method are a good fit to experimental data, but the fit breaks down at higher values of KLaCO2. How to correct KLaCO2 for gas phase enrichment remains to be determined. The one-parameter linear regression model was used to vary the C*CO2 over the test, but it did not result in a better fit to the experimental data when compared to the ASCE or fixed C*CO2 assumptions.

Statistical tools for transgene copy number estimation based on real-time PCR.

PubMed

Yuan, Joshua S; Burris, Jason; Stewart, Nathan R; Mentewab, Ayalew; Stewart, C Neal

2007-11-01

As compared with traditional transgene copy number detection technologies such as Southern blot analysis, real-time PCR provides a fast, inexpensive and high-throughput alternative. However, the real-time PCR based transgene copy number estimation tends to be ambiguous and subjective stemming from the lack of proper statistical analysis and data quality control to render a reliable estimation of copy number with a prediction value. Despite the recent progresses in statistical analysis of real-time PCR, few publications have integrated these advancements in real-time PCR based transgene copy number determination. Three experimental designs and four data quality control integrated statistical models are presented. For the first method, external calibration curves are established for the transgene based on serially-diluted templates. The Ct number from a control transgenic event and putative transgenic event are compared to derive the transgene copy number or zygosity estimation. Simple linear regression and two group T-test procedures were combined to model the data from this design. For the second experimental design, standard curves were generated for both an internal reference gene and the transgene, and the copy number of transgene was compared with that of internal reference gene. Multiple regression models and ANOVA models can be employed to analyze the data and perform quality control for this approach. In the third experimental design, transgene copy number is compared with reference gene without a standard curve, but rather, is based directly on fluorescence data. Two different multiple regression models were proposed to analyze the data based on two different approaches of amplification efficiency integration. Our results highlight the importance of proper statistical treatment and quality control integration in real-time PCR-based transgene copy number determination. These statistical methods allow the real-time PCR-based transgene copy number estimation to be more reliable and precise with a proper statistical estimation. Proper confidence intervals are necessary for unambiguous prediction of trangene copy number. The four different statistical methods are compared for their advantages and disadvantages. Moreover, the statistical methods can also be applied for other real-time PCR-based quantification assays including transfection efficiency analysis and pathogen quantification.

Glomerular structural-functional relationship models of diabetic nephropathy are robust in type 1 diabetic patients.

PubMed

Mauer, Michael; Caramori, Maria Luiza; Fioretto, Paola; Najafian, Behzad

2015-06-01

Studies of structural-functional relationships have improved understanding of the natural history of diabetic nephropathy (DN). However, in order to consider structural end points for clinical trials, the robustness of the resultant models needs to be verified. This study examined whether structural-functional relationship models derived from a large cohort of type 1 diabetic (T1D) patients with a wide range of renal function are robust. The predictability of models derived from multiple regression analysis and piecewise linear regression analysis was also compared. T1D patients (n = 161) with research renal biopsies were divided into two equal groups matched for albumin excretion rate (AER). Models to explain AER and glomerular filtration rate (GFR) by classical DN lesions in one group (T1D-model, or T1D-M) were applied to the other group (T1D-test, or T1D-T) and regression analyses were performed. T1D-M-derived models explained 70 and 63% of AER variance and 32 and 21% of GFR variance in T1D-M and T1D-T, respectively, supporting the substantial robustness of the models. Piecewise linear regression analyses substantially improved predictability of the models with 83% of AER variance and 66% of GFR variance explained by classical DN glomerular lesions alone. These studies demonstrate that DN structural-functional relationship models are robust, and if appropriate models are used, glomerular lesions alone explain a major proportion of AER and GFR variance in T1D patients. © The Author 2014. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.

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.

Linking in situ LAI and fine resolution remote sensing data to map reference LAI over cropland and grassland using geostatistical regression method

NASA Astrophysics Data System (ADS)

He, Yaqian; Bo, Yanchen; Chai, Leilei; Liu, Xiaolong; Li, Aihua

2016-08-01

Leaf Area Index (LAI) is an important parameter of vegetation structure. A number of moderate resolution LAI products have been produced in urgent need of large scale vegetation monitoring. High resolution LAI reference maps are necessary to validate these LAI products. This study used a geostatistical regression (GR) method to estimate LAI reference maps by linking in situ LAI and Landsat TM/ETM+ and SPOT-HRV data over two cropland and two grassland sites. To explore the discrepancies of employing different vegetation indices (VIs) on estimating LAI reference maps, this study established the GR models for different VIs, including difference vegetation index (DVI), normalized difference vegetation index (NDVI), and ratio vegetation index (RVI). To further assess the performance of the GR model, the results from the GR and Reduced Major Axis (RMA) models were compared. The results show that the performance of the GR model varies between the cropland and grassland sites. At the cropland sites, the GR model based on DVI provides the best estimation, while at the grassland sites, the GR model based on DVI performs poorly. Compared to the RMA model, the GR model improves the accuracy of reference LAI maps in terms of root mean square errors (RMSE) and bias.

Comparison of Prediction Model for Cardiovascular Autonomic Dysfunction Using Artificial Neural Network and Logistic Regression Analysis

PubMed Central

Zeng, Fangfang; Li, Zhongtao; Yu, Xiaoling; Zhou, Linuo

2013-01-01

Background This study aimed to develop the artificial neural network (ANN) and multivariable logistic regression (LR) analyses for prediction modeling of cardiovascular autonomic (CA) dysfunction in the general population, and compare the prediction models using the two approaches. Methods and Materials We analyzed a previous dataset based on a Chinese population sample consisting of 2,092 individuals aged 30–80 years. The prediction models were derived from an exploratory set using ANN and LR analysis, and were tested in the validation set. Performances of these prediction models were then compared. Results Univariate analysis indicated that 14 risk factors showed statistically significant association with the prevalence of CA dysfunction (P<0.05). The mean area under the receiver-operating curve was 0.758 (95% CI 0.724–0.793) for LR and 0.762 (95% CI 0.732–0.793) for ANN analysis, but noninferiority result was found (P<0.001). The similar results were found in comparisons of sensitivity, specificity, and predictive values in the prediction models between the LR and ANN analyses. Conclusion The prediction models for CA dysfunction were developed using ANN and LR. ANN and LR are two effective tools for developing prediction models based on our dataset. PMID:23940593

Development of a noise prediction model based on advanced fuzzy approaches in typical industrial workrooms.

PubMed

Aliabadi, Mohsen; Golmohammadi, Rostam; Khotanlou, Hassan; Mansoorizadeh, Muharram; Salarpour, Amir

2014-01-01

Noise prediction is considered to be the best method for evaluating cost-preventative noise controls in industrial workrooms. One of the most important issues is the development of accurate models for analysis of the complex relationships among acoustic features affecting noise level in workrooms. In this study, advanced fuzzy approaches were employed to develop relatively accurate models for predicting noise in noisy industrial workrooms. The data were collected from 60 industrial embroidery workrooms in the Khorasan Province, East of Iran. The main acoustic and embroidery process features that influence the noise were used to develop prediction models using MATLAB software. Multiple regression technique was also employed and its results were compared with those of fuzzy approaches. Prediction errors of all prediction models based on fuzzy approaches were within the acceptable level (lower than one dB). However, Neuro-fuzzy model (RMSE=0.53dB and R2=0.88) could slightly improve the accuracy of noise prediction compared with generate fuzzy model. Moreover, fuzzy approaches provided more accurate predictions than did regression technique. The developed models based on fuzzy approaches as useful prediction tools give professionals the opportunity to have an optimum decision about the effectiveness of acoustic treatment scenarios in embroidery workrooms.

Infection biomarkers in primary care patients with acute respiratory tract infections-comparison of Procalcitonin and C-reactive protein.

PubMed

Meili, Marc; Kutz, Alexander; Briel, Matthias; Christ-Crain, Mirjam; Bucher, Heiner C; Mueller, Beat; Schuetz, Philipp

2016-03-24

There is a lack of studies comparing the utility of C-reactive protein (CRP) with Procalcitonin (PCT) for the management of patients with acute respiratory tract infections (ARI) in primary care. Our aim was to study the correlation between these markers and to compare their predictive accuracy in regard to clinical outcome prediction. This is a secondary analysis using clinical and biomarker data of 458 primary care patients with pneumonic and non-pneumonic ARI. We used correlation statistics (spearman's rank test) and multivariable regression models to assess association of markers with adverse outcome, namely days with restricted activities and persistence of discomfort from infection at day 14. At baseline, CRP and PCT did not correlate well in the overall population (r(2) = 0.16) and particularly in the subgroup of patients with non-pneumonic ARI (r(2) = 0.08). Low correlation of biomarkers were also found when comparing cut-off ranges, day seven levels or changes from baseline to day seven. High baseline levels of CRP (>100 mg/dL, regression coefficient 1.6, 95 % CI 0.5 to 2.6, sociodemographic-adjusted model) as well as PCT (>0.5ug/L regression coefficient 2.0, 95 % CI 0.0 to 4.0, sociodemographic-adjusted model) were significantly associated with larger number of days with restricted activities. There were no associations of either biomarker with persistence of discomfort at day 14. CRP and PCT levels do not well correlate, but both have moderate prognostic accuracy in primary care patients with ARI to predict clinical outcomes. The low correlation between the two biomarkers calls for interventional research comparing these markers head to head in regard to their ability to guide antibiotic decisions. Current Controlled Trials, ISRCTN73182671.

A nonparametric method for assessment of interactions in a median regression model for analyzing right censored data.

PubMed

Lee, MinJae; Rahbar, Mohammad H; Talebi, Hooshang

2018-01-01

We propose a nonparametric test for interactions when we are concerned with investigation of the simultaneous effects of two or more factors in a median regression model with right censored survival data. Our approach is developed to detect interaction in special situations, when the covariates have a finite number of levels with a limited number of observations in each level, and it allows varying levels of variance and censorship at different levels of the covariates. Through simulation studies, we compare the power of detecting an interaction between the study group variable and a covariate using our proposed procedure with that of the Cox Proportional Hazard (PH) model and censored quantile regression model. We also assess the impact of censoring rate and type on the standard error of the estimators of parameters. Finally, we illustrate application of our proposed method to real life data from Prospective Observational Multicenter Major Trauma Transfusion (PROMMTT) study to test an interaction effect between type of injury and study sites using median time for a trauma patient to receive three units of red blood cells. The results from simulation studies indicate that our procedure performs better than both Cox PH model and censored quantile regression model based on statistical power for detecting the interaction, especially when the number of observations is small. It is also relatively less sensitive to censoring rates or even the presence of conditionally independent censoring that is conditional on the levels of covariates.

Accurate Descriptions of Hot Flow Behaviors Across β Transus of Ti-6Al-4V Alloy by Intelligence Algorithm GA-SVR

NASA Astrophysics Data System (ADS)

Wang, Li-yong; Li, Le; Zhang, Zhi-hua

2016-09-01

Hot compression tests of Ti-6Al-4V alloy in a wide temperature range of 1023-1323 K and strain rate range of 0.01-10 s-1 were conducted by a servo-hydraulic and computer-controlled Gleeble-3500 machine. In order to accurately and effectively characterize the highly nonlinear flow behaviors, support vector regression (SVR) which is a machine learning method was combined with genetic algorithm (GA) for characterizing the flow behaviors, namely, the GA-SVR. The prominent character of GA-SVR is that it with identical training parameters will keep training accuracy and prediction accuracy at a stable level in different attempts for a certain dataset. The learning abilities, generalization abilities, and modeling efficiencies of the mathematical regression model, ANN, and GA-SVR for Ti-6Al-4V alloy were detailedly compared. Comparison results show that the learning ability of the GA-SVR is stronger than the mathematical regression model. The generalization abilities and modeling efficiencies of these models were shown as follows in ascending order: the mathematical regression model < ANN < GA-SVR. The stress-strain data outside experimental conditions were predicted by the well-trained GA-SVR, which improved simulation accuracy of the load-stroke curve and can further improve the related research fields where stress-strain data play important roles, such as speculating work hardening and dynamic recovery, characterizing dynamic recrystallization evolution, and improving processing maps.

ATLS Hypovolemic Shock Classification by Prediction of Blood Loss in Rats Using Regression Models.

PubMed

Choi, Soo Beom; Choi, Joon Yul; Park, Jee Soo; Kim, Deok Won

2016-07-01

In our previous study, our input data set consisted of 78 rats, the blood loss in percent as a dependent variable, and 11 independent variables (heart rate, systolic blood pressure, diastolic blood pressure, mean arterial pressure, pulse pressure, respiration rate, temperature, perfusion index, lactate concentration, shock index, and new index (lactate concentration/perfusion)). The machine learning methods for multicategory classification were applied to a rat model in acute hemorrhage to predict the four Advanced Trauma Life Support (ATLS) hypovolemic shock classes for triage in our previous study. However, multicategory classification is much more difficult and complicated than binary classification. We introduce a simple approach for classifying ATLS hypovolaemic shock class by predicting blood loss in percent using support vector regression and multivariate linear regression (MLR). We also compared the performance of the classification models using absolute and relative vital signs. The accuracies of support vector regression and MLR models with relative values by predicting blood loss in percent were 88.5% and 84.6%, respectively. These were better than the best accuracy of 80.8% of the direct multicategory classification using the support vector machine one-versus-one model in our previous study for the same validation data set. Moreover, the simple MLR models with both absolute and relative values could provide possibility of the future clinical decision support system for ATLS classification. The perfusion index and new index were more appropriate with relative changes than absolute values.

Evaluation of the Bitterness of Traditional Chinese Medicines using an E-Tongue Coupled with a Robust Partial Least Squares Regression Method

PubMed Central

Lin, Zhaozhou; Zhang, Qiao; Liu, Ruixin; Gao, Xiaojie; Zhang, Lu; Kang, Bingya; Shi, Junhan; Wu, Zidan; Gui, Xinjing; Li, Xuelin

2016-01-01

To accurately, safely, and efficiently evaluate the bitterness of Traditional Chinese Medicines (TCMs), a robust predictor was developed using robust partial least squares (RPLS) regression method based on data obtained from an electronic tongue (e-tongue) system. The data quality was verified by the Grubb’s test. Moreover, potential outliers were detected based on both the standardized residual and score distance calculated for each sample. The performance of RPLS on the dataset before and after outlier detection was compared to other state-of-the-art methods including multivariate linear regression, least squares support vector machine, and the plain partial least squares regression. Both R2 and root-mean-squares error (RMSE) of cross-validation (CV) were recorded for each model. With four latent variables, a robust RMSECV value of 0.3916 with bitterness values ranging from 0.63 to 4.78 were obtained for the RPLS model that was constructed based on the dataset including outliers. Meanwhile, the RMSECV, which was calculated using the models constructed by other methods, was larger than that of the RPLS model. After six outliers were excluded, the performance of all benchmark methods markedly improved, but the difference between the RPLS model constructed before and after outlier exclusion was negligible. In conclusion, the bitterness of TCM decoctions can be accurately evaluated with the RPLS model constructed using e-tongue data. PMID:26821026

Addressing data privacy in matched studies via virtual pooling.

PubMed

Saha-Chaudhuri, P; Weinberg, C R

2017-09-07

Data confidentiality and shared use of research data are two desirable but sometimes conflicting goals in research with multi-center studies and distributed data. While ideal for straightforward analysis, confidentiality restrictions forbid creation of a single dataset that includes covariate information of all participants. Current approaches such as aggregate data sharing, distributed regression, meta-analysis and score-based methods can have important limitations. We propose a novel application of an existing epidemiologic tool, specimen pooling, to enable confidentiality-preserving analysis of data arising from a matched case-control, multi-center design. Instead of pooling specimens prior to assay, we apply the methodology to virtually pool (aggregate) covariates within nodes. Such virtual pooling retains most of the information used in an analysis with individual data and since individual participant data is not shared externally, within-node virtual pooling preserves data confidentiality. We show that aggregated covariate levels can be used in a conditional logistic regression model to estimate individual-level odds ratios of interest. The parameter estimates from the standard conditional logistic regression are compared to the estimates based on a conditional logistic regression model with aggregated data. The parameter estimates are shown to be similar to those without pooling and to have comparable standard errors and confidence interval coverage. Virtual data pooling can be used to maintain confidentiality of data from multi-center study and can be particularly useful in research with large-scale distributed data.

Introduction to the use of regression models in epidemiology.

PubMed

Bender, Ralf

2009-01-01

Regression modeling is one of the most important statistical techniques used in analytical epidemiology. By means of regression models the effect of one or several explanatory variables (e.g., exposures, subject characteristics, risk factors) on a response variable such as mortality or cancer can be investigated. From multiple regression models, adjusted effect estimates can be obtained that take the effect of potential confounders into account. Regression methods can be applied in all epidemiologic study designs so that they represent a universal tool for data analysis in epidemiology. Different kinds of regression models have been developed in dependence on the measurement scale of the response variable and the study design. The most important methods are linear regression for continuous outcomes, logistic regression for binary outcomes, Cox regression for time-to-event data, and Poisson regression for frequencies and rates. This chapter provides a nontechnical introduction to these regression models with illustrating examples from cancer research.

Bentamapimod (JNK Inhibitor AS602801) Induces Regression of Endometriotic Lesions in Animal Models.

PubMed

Palmer, Stephen S; Altan, Melis; Denis, Deborah; Tos, Enrico Gillio; Gotteland, Jean-Pierre; Osteen, Kevin G; Bruner-Tran, Kaylon L; Nataraja, Selvaraj G

2016-01-01

Endometriosis is an estrogen (ER)-dependent gynecological disease caused by the growth of endometrial tissue at extrauterine sites. Current endocrine therapies address the estrogenic aspect of disease and offer some relief from pain but are associated with significant side effects. Immune dysfunction is also widely believed to be an underlying contributor to the pathogenesis of this disease. This study evaluated an inhibitor of c-Jun N-terminal kinase, bentamapimod (AS602801), which interrupts immune pathways, in 2 rodent endometriosis models. Treatment of nude mice bearing xenografts biopsied from women with endometriosis (BWE) with 30 mg/kg AS602801 caused 29% regression of lesion. Medroxyprogesterone acetate (MPA) or progesterone (PR) alone did not cause regression of BWE lesions, but combining 10 mg/kg AS602801 with MPA caused 38% lesion regression. In human endometrial organ cultures (from healthy women), treatment with AS602801 or MPA reduced matrix metalloproteinase-3 (MMP-3) release into culture medium. In organ cultures established with BWE, PR or MPA failed to inhibit MMP-3 secretion, whereas AS602801 alone or MPA + AS602801 suppressed MMP-3 production. In an autologous rat endometriosis model, AS602801 caused 48% regression of lesions compared to GnRH antagonist Antide (84%). AS602801 reduced inflammatory cytokines in endometriotic lesions, while levels of cytokines in ipsilateral horns were unaffected. Furthermore, AS602801 enhanced natural killer cell activity, without apparent negative effects on uterus. These results indicate that bentamapimod induced regression of endometriotic lesions in endometriosis rodent animal models without suppressing ER action. c-Jun N-terminal kinase inhibition mediated a comprehensive reduction in cytokine secretion and moreover was able to overcome PR resistance. © The Author(s) 2015.

Identifying Interacting Genetic Variations by Fish-Swarm Logic Regression

PubMed Central

Yang, Aiyuan; Yan, Chunxia; Zhu, Feng; Zhao, Zhongmeng; Cao, Zhi

2013-01-01

Understanding associations between genotypes and complex traits is a fundamental problem in human genetics. A major open problem in mapping phenotypes is that of identifying a set of interacting genetic variants, which might contribute to complex traits. Logic regression (LR) is a powerful multivariant association tool. Several LR-based approaches have been successfully applied to different datasets. However, these approaches are not adequate with regard to accuracy and efficiency. In this paper, we propose a new LR-based approach, called fish-swarm logic regression (FSLR), which improves the logic regression process by incorporating swarm optimization. In our approach, a school of fish agents are conducted in parallel. Each fish agent holds a regression model, while the school searches for better models through various preset behaviors. A swarm algorithm improves the accuracy and the efficiency by speeding up the convergence and preventing it from dropping into local optimums. We apply our approach on a real screening dataset and a series of simulation scenarios. Compared to three existing LR-based approaches, our approach outperforms them by having lower type I and type II error rates, being able to identify more preset causal sites, and performing at faster speeds. PMID:23984382

Modeling ionospheric foF 2 response during geomagnetic storms using neural network and linear regression techniques

NASA Astrophysics Data System (ADS)

Tshisaphungo, Mpho; Habarulema, John Bosco; McKinnell, Lee-Anne

2018-06-01

In this paper, the modeling of the ionospheric foF 2 changes during geomagnetic storms by means of neural network (NN) and linear regression (LR) techniques is presented. The results will lead to a valuable tool to model the complex ionospheric changes during disturbed days in an operational space weather monitoring and forecasting environment. The storm-time foF 2 data during 1996-2014 from Grahamstown (33.3°S, 26.5°E), South Africa ionosonde station was used in modeling. In this paper, six storms were reserved to validate the models and hence not used in the modeling process. We found that the performance of both NN and LR models is comparable during selected storms which fell within the data period (1996-2014) used in modeling. However, when validated on storm periods beyond 1996-2014, the NN model gives a better performance (R = 0.62) compared to LR model (R = 0.56) for a storm that reached a minimum Dst index of -155 nT during 19-23 December 2015. We also found that both NN and LR models are capable of capturing the ionospheric foF 2 responses during two great geomagnetic storms (28 October-1 November 2003 and 6-12 November 2004) which have been demonstrated to be difficult storms to model in previous studies.

Comparison of robustness to outliers between robust poisson models and log-binomial models when estimating relative risks for common binary outcomes: a simulation study.

PubMed

Chen, Wansu; Shi, Jiaxiao; Qian, Lei; Azen, Stanley P

2014-06-26

To estimate relative risks or risk ratios for common binary outcomes, the most popular model-based methods are the robust (also known as modified) Poisson and the log-binomial regression. Of the two methods, it is believed that the log-binomial regression yields more efficient estimators because it is maximum likelihood based, while the robust Poisson model may be less affected by outliers. Evidence to support the robustness of robust Poisson models in comparison with log-binomial models is very limited. In this study a simulation was conducted to evaluate the performance of the two methods in several scenarios where outliers existed. The findings indicate that for data coming from a population where the relationship between the outcome and the covariate was in a simple form (e.g. log-linear), the two models yielded comparable biases and mean square errors. However, if the true relationship contained a higher order term, the robust Poisson models consistently outperformed the log-binomial models even when the level of contamination is low. The robust Poisson models are more robust (or less sensitive) to outliers compared to the log-binomial models when estimating relative risks or risk ratios for common binary outcomes. Users should be aware of the limitations when choosing appropriate models to estimate relative risks or risk ratios.

Comparative Analysis of River Flow Modelling by Using Supervised Learning Technique

NASA Astrophysics Data System (ADS)

Ismail, Shuhaida; Mohamad Pandiahi, Siraj; Shabri, Ani; Mustapha, Aida

2018-04-01

The goal of this research is to investigate the efficiency of three supervised learning algorithms for forecasting monthly river flow of the Indus River in Pakistan, spread over 550 square miles or 1800 square kilometres. The algorithms include the Least Square Support Vector Machine (LSSVM), Artificial Neural Network (ANN) and Wavelet Regression (WR). The forecasting models predict the monthly river flow obtained from the three models individually for river flow data and the accuracy of the all models were then compared against each other. The monthly river flow of the said river has been forecasted using these three models. The obtained results were compared and statistically analysed. Then, the results of this analytical comparison showed that LSSVM model is more precise in the monthly river flow forecasting. It was found that LSSVM has he higher r with the value of 0.934 compared to other models. This indicate that LSSVM is more accurate and efficient as compared to the ANN and WR model.

Improving the Spatial Prediction of Soil Organic Carbon Stocks in a Complex Tropical Mountain Landscape by Methodological Specifications in Machine Learning Approaches

PubMed Central

Schmidt, Johannes; Glaser, Bruno

2016-01-01

Tropical forests are significant carbon sinks and their soils’ carbon storage potential is immense. However, little is known about the soil organic carbon (SOC) stocks of tropical mountain areas whose complex soil-landscape and difficult accessibility pose a challenge to spatial analysis. The choice of methodology for spatial prediction is of high importance to improve the expected poor model results in case of low predictor-response correlations. Four aspects were considered to improve model performance in predicting SOC stocks of the organic layer of a tropical mountain forest landscape: Different spatial predictor settings, predictor selection strategies, various machine learning algorithms and model tuning. Five machine learning algorithms: random forests, artificial neural networks, multivariate adaptive regression splines, boosted regression trees and support vector machines were trained and tuned to predict SOC stocks from predictors derived from a digital elevation model and satellite image. Topographical predictors were calculated with a GIS search radius of 45 to 615 m. Finally, three predictor selection strategies were applied to the total set of 236 predictors. All machine learning algorithms—including the model tuning and predictor selection—were compared via five repetitions of a tenfold cross-validation. The boosted regression tree algorithm resulted in the overall best model. SOC stocks ranged between 0.2 to 17.7 kg m-2, displaying a huge variability with diffuse insolation and curvatures of different scale guiding the spatial pattern. Predictor selection and model tuning improved the models’ predictive performance in all five machine learning algorithms. The rather low number of selected predictors favours forward compared to backward selection procedures. Choosing predictors due to their indiviual performance was vanquished by the two procedures which accounted for predictor interaction. PMID:27128736

Patient satisfaction with ambulatory care in Germany: effects of patient- and medical practice-related factors.

PubMed

Auras, Silke; Ostermann, Thomas; de Cruppé, Werner; Bitzer, Eva-Maria; Diel, Franziska; Geraedts, Max

2016-12-01

The study aimed to illustrate the effect of the patients' sex, age, self-rated health and medical practice specialization on patient satisfaction. Secondary analysis of patient survey data using multilevel analysis (generalized linear mixed model, medical practice as random effect) using a sequential modelling strategy. We examined the effects of the patients' sex, age, self-rated health and medical practice specialization on four patient satisfaction dimensions: medical practice organization, information, interaction, professional competence. The study was performed in 92 German medical practices providing ambulatory care in general medicine, internal medicine or gynaecology. In total, 9888 adult patients participated in a patient survey using the validated 'questionnaire on satisfaction with ambulatory care-quality from the patient perspective [ZAP]'. We calculated four models for each satisfaction dimension, revealing regression coefficients with 95% confidence intervals (CIs) for all independent variables, and using Wald Chi-Square statistic for each modelling step (model validity) and LR-Tests to compare the models of each step with the previous model. The patients' sex and age had a weak effect (maximum regression coefficient 1.09, CI 0.39; 1.80), and the patients' self-rated health had the strongest positive effect (maximum regression coefficient 7.66, CI 6.69; 8.63) on satisfaction ratings. The effect of medical practice specialization was heterogeneous. All factors studied, specifically the patients' self-rated health, affected patient satisfaction. Adjustment should always be considered because it improves the comparability of patient satisfaction in medical practices with atypically varying patient populations and increases the acceptance of comparisons. © The Author 2016. Published by Oxford University Press in association with the International Society for Quality in Health Care. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Assessing the blood volume and heart rate responses during haemodialysis in fluid overloaded patients using support vector regression.

PubMed

Javed, Faizan; Savkin, Andrey V; Chan, Gregory S H; Middleton, Paul M; Malouf, Philip; Steel, Elizabeth; Mackie, James; Lovell, Nigel H

2009-11-01

This study aims to assess the blood volume and heart rate (HR) responses during haemodialysis in fluid overloaded patients by a nonparametric nonlinear regression approach based on a support vector machine (SVM). Relative blood volume (RBV) and electrocardiogram (ECG) was recorded from 23 haemodynamically stable renal failure patients during regular haemodialysis. Modelling was performed on 18 fluid overloaded patients (fluid removal of >2 L). SVM-based regression was used to obtain the models of RBV change with time as well as the percentage change in HR with respect to RBV. Mean squared error (MSE) and goodness of fit (R(2)) were used for comparison among different kernel functions. The design parameters were estimated using a grid search approach and the selected models were validated by a k-fold cross-validation technique. For the model of HR versus RBV change, a radial basis function (RBF) kernel (MSE = 17.37 and R(2) = 0.932) gave the least MSE compared to linear (MSE = 25.97 and R(2) = 0.898) and polynomial (MSE = 18.18 and R(2)= 0.929). The MSE was significantly lower for training data set when using RBF kernel compared to other kernels (p < 0.01). The RBF kernel also provided a slightly better fit of RBV change with time (MSE = 1.12 and R(2) = 0.91) compared to a linear kernel (MSE = 1.46 and R(2) = 0.88). The modelled HR response was characterized by an initial drop and a subsequent rise during progressive reduction in RBV, which may be interpreted as the reflex response to a transition from central hypervolaemia to hypovolaemia. These modelled curves can be used as references to a controller that can be designed to regulate the haemodynamic variables to ensure the stability of patients undergoing haemodialysis.

How is the weather? Forecasting inpatient glycemic control

PubMed Central

Saulnier, George E; Castro, Janna C; Cook, Curtiss B; Thompson, Bithika M

2017-01-01

Aim: Apply methods of damped trend analysis to forecast inpatient glycemic control. Method: Observed and calculated point-of-care blood glucose data trends were determined over 62 weeks. Mean absolute percent error was used to calculate differences between observed and forecasted values. Comparisons were drawn between model results and linear regression forecasting. Results: The forecasted mean glucose trends observed during the first 24 and 48 weeks of projections compared favorably to the results provided by linear regression forecasting. However, in some scenarios, the damped trend method changed inferences compared with linear regression. In all scenarios, mean absolute percent error values remained below the 10% accepted by demand industries. Conclusion: Results indicate that forecasting methods historically applied within demand industries can project future inpatient glycemic control. Additional study is needed to determine if forecasting is useful in the analyses of other glucometric parameters and, if so, how to apply the techniques to quality improvement. PMID:29134125

Harmony search optimization in dimensional accuracy of die sinking EDM process using SS316L stainless steel

NASA Astrophysics Data System (ADS)

Deris, A. M.; Zain, A. M.; Sallehuddin, R.; Sharif, S.

2017-09-01

Electric discharge machine (EDM) is one of the widely used nonconventional machining processes for hard and difficult to machine materials. Due to the large number of machining parameters in EDM and its complicated structural, the selection of the optimal solution of machining parameters for obtaining minimum machining performance is remain as a challenging task to the researchers. This paper proposed experimental investigation and optimization of machining parameters for EDM process on stainless steel 316L work piece using Harmony Search (HS) algorithm. The mathematical model was developed based on regression approach with four input parameters which are pulse on time, peak current, servo voltage and servo speed to the output response which is dimensional accuracy (DA). The optimal result of HS approach was compared with regression analysis and it was found HS gave better result y giving the most minimum DA value compared with regression approach.

Hypothesis Testing Using Factor Score Regression

PubMed Central

Devlieger, Ines; Mayer, Axel; Rosseel, Yves

2015-01-01

In this article, an overview is given of four methods to perform factor score regression (FSR), namely regression FSR, Bartlett FSR, the bias avoiding method of Skrondal and Laake, and the bias correcting method of Croon. The bias correcting method is extended to include a reliable standard error. The four methods are compared with each other and with structural equation modeling (SEM) by using analytic calculations and two Monte Carlo simulation studies to examine their finite sample characteristics. Several performance criteria are used, such as the bias using the unstandardized and standardized parameterization, efficiency, mean square error, standard error bias, type I error rate, and power. The results show that the bias correcting method, with the newly developed standard error, is the only suitable alternative for SEM. While it has a higher standard error bias than SEM, it has a comparable bias, efficiency, mean square error, power, and type I error rate. PMID:29795886

Decline in bloater fecundity in Southern Lake Michigan after decline of Diporeia

USGS Publications Warehouse

Bunnell, D.B.; David, S.R.; Madenjian, C.P.

2009-01-01

Population fecundity can vary through time, sometimes owing to changes in adult condition. Consideration of these fecundity changes can improve understanding of recruitment variation. Herein, we estimated fecundity of Lake Michigan bloater Coregonus hoyi during December 2005 and February 2006. Bloater recruitment has been highly variable from 1962 to present, and consistently poor since 1992. We compared our fecundity vs. weight regression to a previously published regression that used fish sampled in October 1969. We wanted to develop a new regression for two reasons. First, it should be more accurate because it uses fish collected closer to spawning, thus minimizing the potential for atresia (egg reabsorption) which could bias fecundity high. Second, we hypothesized that fecundity would be lower in 2006 because adult condition was 41% lower in 2006 compared to 1969, likely owing to the decline of Diporeia spp, a primary prey for bloater. Although the slope of the fecundity versus weight regression was similar between the years, fecundity was 24% lower in 2006 than in 1969 for bloater weighing between 70 and 240??g. Whether this was the result of the difference in sampling time prior to spawning or of differences in condition is unknown. We also found no relationship between maternal size and mature oocyte size. Incorporating our updated fecundity regression into a stock/recruit model failed to improve the model fit, indicating that the low bloater recruitment that has been observed since the early 1990s is not solely the result of reduced fecundity. ?? 2008 Elsevier Inc. All rights reserved.

Healthcare Expenditures Associated with Depression Among Individuals with Osteoarthritis: Post-Regression Linear Decomposition Approach.

PubMed

Agarwal, Parul; Sambamoorthi, Usha

2015-12-01

Depression is common among individuals with osteoarthritis and leads to increased healthcare burden. The objective of this study was to examine excess total healthcare expenditures associated with depression among individuals with osteoarthritis in the US. Adults with self-reported osteoarthritis (n = 1881) were identified using data from the 2010 Medical Expenditure Panel Survey (MEPS). Among those with osteoarthritis, chi-square tests and ordinary least square regressions (OLS) were used to examine differences in healthcare expenditures between those with and without depression. Post-regression linear decomposition technique was used to estimate the relative contribution of different constructs of the Anderson's behavioral model, i.e., predisposing, enabling, need, personal healthcare practices, and external environment factors, to the excess expenditures associated with depression among individuals with osteoarthritis. All analysis accounted for the complex survey design of MEPS. Depression coexisted among 20.6 % of adults with osteoarthritis. The average total healthcare expenditures were $13,684 among adults with depression compared to $9284 among those without depression. Multivariable OLS regression revealed that adults with depression had 38.8 % higher healthcare expenditures (p < 0.001) compared to those without depression. Post-regression linear decomposition analysis indicated that 50 % of differences in expenditures among adults with and without depression can be explained by differences in need factors. Among individuals with coexisting osteoarthritis and depression, excess healthcare expenditures associated with depression were mainly due to comorbid anxiety, chronic conditions and poor health status. These expenditures may potentially be reduced by providing timely intervention for need factors or by providing care under a collaborative care model.

Assessing Hospital Performance After Percutaneous Coronary Intervention Using Big Data.

PubMed

Spertus, Jacob V; T Normand, Sharon-Lise; Wolf, Robert; Cioffi, Matt; Lovett, Ann; Rose, Sherri

2016-11-01

Although risk adjustment remains a cornerstone for comparing outcomes across hospitals, optimal strategies continue to evolve in the presence of many confounders. We compared conventional regression-based model to approaches particularly suited to leveraging big data. We assessed hospital all-cause 30-day excess mortality risk among 8952 adults undergoing percutaneous coronary intervention between October 1, 2011, and September 30, 2012, in 24 Massachusetts hospitals using clinical registry data linked with billing data. We compared conventional logistic regression models with augmented inverse probability weighted estimators and targeted maximum likelihood estimators to generate more efficient and unbiased estimates of hospital effects. We also compared a clinically informed and a machine-learning approach to confounder selection, using elastic net penalized regression in the latter case. Hospital excess risk estimates range from -1.4% to 2.0% across methods and confounder sets. Some hospitals were consistently classified as low or as high excess mortality outliers; others changed classification depending on the method and confounder set used. Switching from the clinically selected list of 11 confounders to a full set of 225 confounders increased the estimation uncertainty by an average of 62% across methods as measured by confidence interval length. Agreement among methods ranged from fair, with a κ statistic of 0.39 (SE: 0.16), to perfect, with a κ of 1 (SE: 0.0). Modern causal inference techniques should be more frequently adopted to leverage big data while minimizing bias in hospital performance assessments. © 2016 American Heart Association, Inc.

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.

Modelling daily dissolved oxygen concentration using least square support vector machine, multivariate adaptive regression splines and M5 model tree

NASA Astrophysics Data System (ADS)

Heddam, Salim; Kisi, Ozgur

2018-04-01

In the present study, three types of artificial intelligence techniques, least square support vector machine (LSSVM), multivariate adaptive regression splines (MARS) and M5 model tree (M5T) are applied for modeling daily dissolved oxygen (DO) concentration using several water quality variables as inputs. The DO concentration and water quality variables data from three stations operated by the United States Geological Survey (USGS) were used for developing the three models. The water quality data selected consisted of daily measured of water temperature (TE, °C), pH (std. unit), specific conductance (SC, μS/cm) and discharge (DI cfs), are used as inputs to the LSSVM, MARS and M5T models. The three models were applied for each station separately and compared to each other. According to the results obtained, it was found that: (i) the DO concentration could be successfully estimated using the three models and (ii) the best model among all others differs from one station to another.