Optimal methods for fitting probability distributions to propagule retention time in studies of zoochorous dispersal.

PubMed

Viana, Duarte S; Santamaría, Luis; Figuerola, Jordi

2016-02-01

Propagule retention time is a key factor in determining propagule dispersal distance and the shape of "seed shadows". Propagules dispersed by animal vectors are either ingested and retained in the gut until defecation or attached externally to the body until detachment. Retention time is a continuous variable, but it is commonly measured at discrete time points, according to pre-established sampling time-intervals. Although parametric continuous distributions have been widely fitted to these interval-censored data, the performance of different fitting methods has not been evaluated. To investigate the performance of five different fitting methods, we fitted parametric probability distributions to typical discretized retention-time data with known distribution using as data-points either the lower, mid or upper bounds of sampling intervals, as well as the cumulative distribution of observed values (using either maximum likelihood or non-linear least squares for parameter estimation); then compared the estimated and original distributions to assess the accuracy of each method. We also assessed the robustness of these methods to variations in the sampling procedure (sample size and length of sampling time-intervals). Fittings to the cumulative distribution performed better for all types of parametric distributions (lognormal, gamma and Weibull distributions) and were more robust to variations in sample size and sampling time-intervals. These estimated distributions had negligible deviations of up to 0.045 in cumulative probability of retention times (according to the Kolmogorov-Smirnov statistic) in relation to original distributions from which propagule retention time was simulated, supporting the overall accuracy of this fitting method. In contrast, fitting the sampling-interval bounds resulted in greater deviations that ranged from 0.058 to 0.273 in cumulative probability of retention times, which may introduce considerable biases in parameter estimates. We recommend the use of cumulative probability to fit parametric probability distributions to propagule retention time, specifically using maximum likelihood for parameter estimation. Furthermore, the experimental design for an optimal characterization of unimodal propagule retention time should contemplate at least 500 recovered propagules and sampling time-intervals not larger than the time peak of propagule retrieval, except in the tail of the distribution where broader sampling time-intervals may also produce accurate fits.

Cure rate model with interval censored data.

PubMed

Kim, Yang-Jin; Jhun, Myoungshic

2008-01-15

In cancer trials, a significant fraction of patients can be cured, that is, the disease is completely eliminated, so that it never recurs. In general, treatments are developed to both increase the patients' chances of being cured and prolong the survival time among non-cured patients. A cure rate model represents a combination of cure fraction and survival model, and can be applied to many clinical studies over several types of cancer. In this article, the cure rate model is considered in the interval censored data composed of two time points, which include the event time of interest. Interval censored data commonly occur in the studies of diseases that often progress without symptoms, requiring clinical evaluation for detection (Encyclopedia of Biostatistics. Wiley: New York, 1998; 2090-2095). In our study, an approximate likelihood approach suggested by Goetghebeur and Ryan (Biometrics 2000; 56:1139-1144) is used to derive the likelihood in interval censored data. In addition, a frailty model is introduced to characterize the association between the cure fraction and survival model. In particular, the positive association between the cure fraction and the survival time is incorporated by imposing a common normal frailty effect. The EM algorithm is used to estimate parameters and a multiple imputation based on the profile likelihood is adopted for variance estimation. The approach is applied to the smoking cessation study in which the event of interest is a smoking relapse and several covariates including an intensive care treatment are evaluated to be effective for both the occurrence of relapse and the non-smoking duration. Copyright (c) 2007 John Wiley & Sons, Ltd.

Abstract: Inference and Interval Estimation for Indirect Effects With Latent Variable Models.

PubMed

Falk, Carl F; Biesanz, Jeremy C

2011-11-30

Models specifying indirect effects (or mediation) and structural equation modeling are both popular in the social sciences. Yet relatively little research has compared methods that test for indirect effects among latent variables and provided precise estimates of the effectiveness of different methods. This simulation study provides an extensive comparison of methods for constructing confidence intervals and for making inferences about indirect effects with latent variables. We compared the percentile (PC) bootstrap, bias-corrected (BC) bootstrap, bias-corrected accelerated (BC a ) bootstrap, likelihood-based confidence intervals (Neale & Miller, 1997), partial posterior predictive (Biesanz, Falk, and Savalei, 2010), and joint significance tests based on Wald tests or likelihood ratio tests. All models included three reflective latent variables representing the independent, dependent, and mediating variables. The design included the following fully crossed conditions: (a) sample size: 100, 200, and 500; (b) number of indicators per latent variable: 3 versus 5; (c) reliability per set of indicators: .7 versus .9; (d) and 16 different path combinations for the indirect effect (α = 0, .14, .39, or .59; and β = 0, .14, .39, or .59). Simulations were performed using a WestGrid cluster of 1680 3.06GHz Intel Xeon processors running R and OpenMx. Results based on 1,000 replications per cell and 2,000 resamples per bootstrap method indicated that the BC and BC a bootstrap methods have inflated Type I error rates. Likelihood-based confidence intervals and the PC bootstrap emerged as methods that adequately control Type I error and have good coverage rates.

Assessing Interval Estimation Methods for Hill Model ...

EPA Pesticide Factsheets

The Hill model of concentration-response is ubiquitous in toxicology, perhaps because its parameters directly relate to biologically significant metrics of toxicity such as efficacy and potency. Point estimates of these parameters obtained through least squares regression or maximum likelihood are commonly used in high-throughput risk assessment, but such estimates typically fail to include reliable information concerning confidence in (or precision of) the estimates. To address this issue, we examined methods for assessing uncertainty in Hill model parameter estimates derived from concentration-response data. In particular, using a sample of ToxCast concentration-response data sets, we applied four methods for obtaining interval estimates that are based on asymptotic theory, bootstrapping (two varieties), and Bayesian parameter estimation, and then compared the results. These interval estimation methods generally did not agree, so we devised a simulation study to assess their relative performance. We generated simulated data by constructing four statistical error models capable of producing concentration-response data sets comparable to those observed in ToxCast. We then applied the four interval estimation methods to the simulated data and compared the actual coverage of the interval estimates to the nominal coverage (e.g., 95%) in order to quantify performance of each of the methods in a variety of cases (i.e., different values of the true Hill model paramet

Statistical inference based on the nonparametric maximum likelihood estimator under double-truncation.

PubMed

Emura, Takeshi; Konno, Yoshihiko; Michimae, Hirofumi

2015-07-01

Doubly truncated data consist of samples whose observed values fall between the right- and left- truncation limits. With such samples, the distribution function of interest is estimated using the nonparametric maximum likelihood estimator (NPMLE) that is obtained through a self-consistency algorithm. Owing to the complicated asymptotic distribution of the NPMLE, the bootstrap method has been suggested for statistical inference. This paper proposes a closed-form estimator for the asymptotic covariance function of the NPMLE, which is computationally attractive alternative to bootstrapping. Furthermore, we develop various statistical inference procedures, such as confidence interval, goodness-of-fit tests, and confidence bands to demonstrate the usefulness of the proposed covariance estimator. Simulations are performed to compare the proposed method with both the bootstrap and jackknife methods. The methods are illustrated using the childhood cancer dataset.

WEIGHTED LIKELIHOOD ESTIMATION UNDER TWO-PHASE SAMPLING

PubMed Central

Saegusa, Takumi; Wellner, Jon A.

2013-01-01

We develop asymptotic theory for weighted likelihood estimators (WLE) under two-phase stratified sampling without replacement. We also consider several variants of WLEs involving estimated weights and calibration. A set of empirical process tools are developed including a Glivenko–Cantelli theorem, a theorem for rates of convergence of M-estimators, and a Donsker theorem for the inverse probability weighted empirical processes under two-phase sampling and sampling without replacement at the second phase. Using these general results, we derive asymptotic distributions of the WLE of a finite-dimensional parameter in a general semiparametric model where an estimator of a nuisance parameter is estimable either at regular or nonregular rates. We illustrate these results and methods in the Cox model with right censoring and interval censoring. We compare the methods via their asymptotic variances under both sampling without replacement and the more usual (and easier to analyze) assumption of Bernoulli sampling at the second phase. PMID:24563559

Time-resolved speckle effects on the estimation of laser-pulse arrival times

NASA Technical Reports Server (NTRS)

Tsai, B.-M.; Gardner, C. S.

1985-01-01

A maximum-likelihood (ML) estimator of the pulse arrival in laser ranging and altimetry is derived for the case of a pulse distorted by shot noise and time-resolved speckle. The performance of the estimator is evaluated for pulse reflections from flat diffuse targets and compared with the performance of a suboptimal centroid estimator and a suboptimal Bar-David ML estimator derived under the assumption of no speckle. In the large-signal limit the accuracy of the estimator was found to improve as the width of the receiver observational interval increases. The timing performance of the estimator is expected to be highly sensitive to background noise when the received pulse energy is high and the receiver observational interval is large. Finally, in the speckle-limited regime the ML estimator performs considerably better than the suboptimal estimators.

Outcome-Dependent Sampling with Interval-Censored Failure Time Data

PubMed Central

Zhou, Qingning; Cai, Jianwen; Zhou, Haibo

2017-01-01

Summary Epidemiologic studies and disease prevention trials often seek to relate an exposure variable to a failure time that suffers from interval-censoring. When the failure rate is low and the time intervals are wide, a large cohort is often required so as to yield reliable precision on the exposure-failure-time relationship. However, large cohort studies with simple random sampling could be prohibitive for investigators with a limited budget, especially when the exposure variables are expensive to obtain. Alternative cost-effective sampling designs and inference procedures are therefore desirable. We propose an outcome-dependent sampling (ODS) design with interval-censored failure time data, where we enrich the observed sample by selectively including certain more informative failure subjects. We develop a novel sieve semiparametric maximum empirical likelihood approach for fitting the proportional hazards model to data from the proposed interval-censoring ODS design. This approach employs the empirical likelihood and sieve methods to deal with the infinite-dimensional nuisance parameters, which greatly reduces the dimensionality of the estimation problem and eases the computation difficulty. The consistency and asymptotic normality of the resulting regression parameter estimator are established. The results from our extensive simulation study show that the proposed design and method works well for practical situations and is more efficient than the alternative designs and competing approaches. An example from the Atherosclerosis Risk in Communities (ARIC) study is provided for illustration. PMID:28771664

Likelihood ratio meta-analysis: New motivation and approach for an old method.

PubMed

Dormuth, Colin R; Filion, Kristian B; Platt, Robert W

2016-03-01

A 95% confidence interval (CI) in an updated meta-analysis may not have the expected 95% coverage. If a meta-analysis is simply updated with additional data, then the resulting 95% CI will be wrong because it will not have accounted for the fact that the earlier meta-analysis failed or succeeded to exclude the null. This situation can be avoided by using the likelihood ratio (LR) as a measure of evidence that does not depend on type-1 error. We show how an LR-based approach, first advanced by Goodman, can be used in a meta-analysis to pool data from separate studies to quantitatively assess where the total evidence points. The method works by estimating the log-likelihood ratio (LogLR) function from each study. Those functions are then summed to obtain a combined function, which is then used to retrieve the total effect estimate, and a corresponding 'intrinsic' confidence interval. Using as illustrations the CAPRIE trial of clopidogrel versus aspirin in the prevention of ischemic events, and our own meta-analysis of higher potency statins and the risk of acute kidney injury, we show that the LR-based method yields the same point estimate as the traditional analysis, but with an intrinsic confidence interval that is appropriately wider than the traditional 95% CI. The LR-based method can be used to conduct both fixed effect and random effects meta-analyses, it can be applied to old and new meta-analyses alike, and results can be presented in a format that is familiar to a meta-analytic audience. Copyright © 2016 Elsevier Inc. All rights reserved.

Empirical Bayes Gaussian likelihood estimation of exposure distributions from pooled samples in human biomonitoring.

PubMed

Li, Xiang; Kuk, Anthony Y C; Xu, Jinfeng

2014-12-10

Human biomonitoring of exposure to environmental chemicals is important. Individual monitoring is not viable because of low individual exposure level or insufficient volume of materials and the prohibitive cost of taking measurements from many subjects. Pooling of samples is an efficient and cost-effective way to collect data. Estimation is, however, complicated as individual values within each pool are not observed but are only known up to their average or weighted average. The distribution of such averages is intractable when the individual measurements are lognormally distributed, which is a common assumption. We propose to replace the intractable distribution of the pool averages by a Gaussian likelihood to obtain parameter estimates. If the pool size is large, this method produces statistically efficient estimates, but regardless of pool size, the method yields consistent estimates as the number of pools increases. An empirical Bayes (EB) Gaussian likelihood approach, as well as its Bayesian analog, is developed to pool information from various demographic groups by using a mixed-effect formulation. We also discuss methods to estimate the underlying mean-variance relationship and to select a good model for the means, which can be incorporated into the proposed EB or Bayes framework. By borrowing strength across groups, the EB estimator is more efficient than the individual group-specific estimator. Simulation results show that the EB Gaussian likelihood estimates outperform a previous method proposed for the National Health and Nutrition Examination Surveys with much smaller bias and better coverage in interval estimation, especially after correction of bias. Copyright © 2014 John Wiley & Sons, Ltd.

Forecasting overhaul or replacement intervals based on estimated system failure intensity

NASA Astrophysics Data System (ADS)

Gannon, James M.

1994-12-01

System reliability can be expressed in terms of the pattern of failure events over time. Assuming a nonhomogeneous Poisson process and Weibull intensity function for complex repairable system failures, the degree of system deterioration can be approximated. Maximum likelihood estimators (MLE's) for the system Rate of Occurrence of Failure (ROCOF) function are presented. Evaluating the integral of the ROCOF over annual usage intervals yields the expected number of annual system failures. By associating a cost of failure with the expected number of failures, budget and program policy decisions can be made based on expected future maintenance costs. Monte Carlo simulation is used to estimate the range and the distribution of the net present value and internal rate of return of alternative cash flows based on the distributions of the cost inputs and confidence intervals of the MLE's.

Interval Estimation of Seismic Hazard Parameters

NASA Astrophysics Data System (ADS)

Orlecka-Sikora, Beata; Lasocki, Stanislaw

2017-03-01

The paper considers Poisson temporal occurrence of earthquakes and presents a way to integrate uncertainties of the estimates of mean activity rate and magnitude cumulative distribution function in the interval estimation of the most widely used seismic hazard functions, such as the exceedance probability and the mean return period. The proposed algorithm can be used either when the Gutenberg-Richter model of magnitude distribution is accepted or when the nonparametric estimation is in use. When the Gutenberg-Richter model of magnitude distribution is used the interval estimation of its parameters is based on the asymptotic normality of the maximum likelihood estimator. When the nonparametric kernel estimation of magnitude distribution is used, we propose the iterated bias corrected and accelerated method for interval estimation based on the smoothed bootstrap and second-order bootstrap samples. The changes resulted from the integrated approach in the interval estimation of the seismic hazard functions with respect to the approach, which neglects the uncertainty of the mean activity rate estimates have been studied using Monte Carlo simulations and two real dataset examples. The results indicate that the uncertainty of mean activity rate affects significantly the interval estimates of hazard functions only when the product of activity rate and the time period, for which the hazard is estimated, is no more than 5.0. When this product becomes greater than 5.0, the impact of the uncertainty of cumulative distribution function of magnitude dominates the impact of the uncertainty of mean activity rate in the aggregated uncertainty of the hazard functions. Following, the interval estimates with and without inclusion of the uncertainty of mean activity rate converge. The presented algorithm is generic and can be applied also to capture the propagation of uncertainty of estimates, which are parameters of a multiparameter function, onto this function.

Intervals for posttest probabilities: a comparison of 5 methods.

PubMed

Mossman, D; Berger, J O

2001-01-01

Several medical articles discuss methods of constructing confidence intervals for single proportions and the likelihood ratio, but scant attention has been given to the systematic study of intervals for the posterior odds, or the positive predictive value, of a test. The authors describe 5 methods of constructing confidence intervals for posttest probabilities when estimates of sensitivity, specificity, and the pretest probability of a disorder are derived from empirical data. They then evaluate each method to determine how well the intervals' coverage properties correspond to their nominal value. When the estimates of pretest probabilities, sensitivity, and specificity are derived from more than 80 subjects and are not close to 0 or 1, all methods generate intervals with appropriate coverage properties. When these conditions are not met, however, the best-performing method is an objective Bayesian approach implemented by a simple simulation using a spreadsheet. Physicians and investigators can generate accurate confidence intervals for posttest probabilities in small-sample situations using the objective Bayesian approach.

Stochastic control system parameter identifiability

NASA Technical Reports Server (NTRS)

Lee, C. H.; Herget, C. J.

1975-01-01

The parameter identification problem of general discrete time, nonlinear, multiple input/multiple output dynamic systems with Gaussian white distributed measurement errors is considered. The knowledge of the system parameterization was assumed to be known. Concepts of local parameter identifiability and local constrained maximum likelihood parameter identifiability were established. A set of sufficient conditions for the existence of a region of parameter identifiability was derived. A computation procedure employing interval arithmetic was provided for finding the regions of parameter identifiability. If the vector of the true parameters is locally constrained maximum likelihood (CML) identifiable, then with probability one, the vector of true parameters is a unique maximal point of the maximum likelihood function in the region of parameter identifiability and the constrained maximum likelihood estimation sequence will converge to the vector of true parameters.

A decision directed detector for the phase incoherent Gaussian channel

NASA Technical Reports Server (NTRS)

Kazakos, D.

1975-01-01

A vector digital signalling scheme is proposed for simultaneous adaptive data transmission and phase estimation. The use of maximum likelihood estimation methods predicts a better performance than the phase-locked loop. The phase estimate is shown to converge to the true value, so that the adaptive nature of the detector effectively achieves phase acquisition and improvement in performance. No separate synchronization interval is required and phase fluctuations can be tracked simultaneously with the transmission of information.

Regression analysis of case K interval-censored failure time data in the presence of informative censoring.

PubMed

Wang, Peijie; Zhao, Hui; Sun, Jianguo

2016-12-01

Interval-censored failure time data occur in many fields such as demography, economics, medical research, and reliability and many inference procedures on them have been developed (Sun, 2006; Chen, Sun, and Peace, 2012). However, most of the existing approaches assume that the mechanism that yields interval censoring is independent of the failure time of interest and it is clear that this may not be true in practice (Zhang et al., 2007; Ma, Hu, and Sun, 2015). In this article, we consider regression analysis of case K interval-censored failure time data when the censoring mechanism may be related to the failure time of interest. For the problem, an estimated sieve maximum-likelihood approach is proposed for the data arising from the proportional hazards frailty model and for estimation, a two-step procedure is presented. In the addition, the asymptotic properties of the proposed estimators of regression parameters are established and an extensive simulation study suggests that the method works well. Finally, we apply the method to a set of real interval-censored data that motivated this study. © 2016, The International Biometric Society.

The Applicability of Confidence Intervals of Quantiles for the Generalized Logistic Distribution

NASA Astrophysics Data System (ADS)

Shin, H.; Heo, J.; Kim, T.; Jung, Y.

2007-12-01

The generalized logistic (GL) distribution has been widely used for frequency analysis. However, there is a little study related to the confidence intervals that indicate the prediction accuracy of distribution for the GL distribution. In this paper, the estimation of the confidence intervals of quantiles for the GL distribution is presented based on the method of moments (MOM), maximum likelihood (ML), and probability weighted moments (PWM) and the asymptotic variances of each quantile estimator are derived as functions of the sample sizes, return periods, and parameters. Monte Carlo simulation experiments are also performed to verify the applicability of the derived confidence intervals of quantile. As the results, the relative bias (RBIAS) and relative root mean square error (RRMSE) of the confidence intervals generally increase as return period increases and reverse as sample size increases. And PWM for estimating the confidence intervals performs better than the other methods in terms of RRMSE when the data is almost symmetric while ML shows the smallest RBIAS and RRMSE when the data is more skewed and sample size is moderately large. The GL model was applied to fit the distribution of annual maximum rainfall data. The results show that there are little differences in the estimated quantiles between ML and PWM while distinct differences in MOM.

Estimation of submarine mass failure probability from a sequence of deposits with age dates

USGS Publications Warehouse

Geist, Eric L.; Chaytor, Jason D.; Parsons, Thomas E.; ten Brink, Uri S.

2013-01-01

The empirical probability of submarine mass failure is quantified from a sequence of dated mass-transport deposits. Several different techniques are described to estimate the parameters for a suite of candidate probability models. The techniques, previously developed for analyzing paleoseismic data, include maximum likelihood and Type II (Bayesian) maximum likelihood methods derived from renewal process theory and Monte Carlo methods. The estimated mean return time from these methods, unlike estimates from a simple arithmetic mean of the center age dates and standard likelihood methods, includes the effects of age-dating uncertainty and of open time intervals before the first and after the last event. The likelihood techniques are evaluated using Akaike’s Information Criterion (AIC) and Akaike’s Bayesian Information Criterion (ABIC) to select the optimal model. The techniques are applied to mass transport deposits recorded in two Integrated Ocean Drilling Program (IODP) drill sites located in the Ursa Basin, northern Gulf of Mexico. Dates of the deposits were constrained by regional bio- and magnetostratigraphy from a previous study. Results of the analysis indicate that submarine mass failures in this location occur primarily according to a Poisson process in which failures are independent and return times follow an exponential distribution. However, some of the model results suggest that submarine mass failures may occur quasiperiodically at one of the sites (U1324). The suite of techniques described in this study provides quantitative probability estimates of submarine mass failure occurrence, for any number of deposits and age uncertainty distributions.

Bayesian forecasting and uncertainty quantifying of stream flows using Metropolis-Hastings Markov Chain Monte Carlo algorithm

NASA Astrophysics Data System (ADS)

Wang, Hongrui; Wang, Cheng; Wang, Ying; Gao, Xiong; Yu, Chen

2017-06-01

This paper presents a Bayesian approach using Metropolis-Hastings Markov Chain Monte Carlo algorithm and applies this method for daily river flow rate forecast and uncertainty quantification for Zhujiachuan River using data collected from Qiaotoubao Gage Station and other 13 gage stations in Zhujiachuan watershed in China. The proposed method is also compared with the conventional maximum likelihood estimation (MLE) for parameter estimation and quantification of associated uncertainties. While the Bayesian method performs similarly in estimating the mean value of daily flow rate, it performs over the conventional MLE method on uncertainty quantification, providing relatively narrower reliable interval than the MLE confidence interval and thus more precise estimation by using the related information from regional gage stations. The Bayesian MCMC method might be more favorable in the uncertainty analysis and risk management.

Likelihood inference for COM-Poisson cure rate model with interval-censored data and Weibull lifetimes.

PubMed

Pal, Suvra; Balakrishnan, N

2017-10-01

In this paper, we consider a competing cause scenario and assume the number of competing causes to follow a Conway-Maxwell Poisson distribution which can capture both over and under dispersion that is usually encountered in discrete data. Assuming the population of interest having a component cure and the form of the data to be interval censored, as opposed to the usually considered right-censored data, the main contribution is in developing the steps of the expectation maximization algorithm for the determination of the maximum likelihood estimates of the model parameters of the flexible Conway-Maxwell Poisson cure rate model with Weibull lifetimes. An extensive Monte Carlo simulation study is carried out to demonstrate the performance of the proposed estimation method. Model discrimination within the Conway-Maxwell Poisson distribution is addressed using the likelihood ratio test and information-based criteria to select a suitable competing cause distribution that provides the best fit to the data. A simulation study is also carried out to demonstrate the loss in efficiency when selecting an improper competing cause distribution which justifies the use of a flexible family of distributions for the number of competing causes. Finally, the proposed methodology and the flexibility of the Conway-Maxwell Poisson distribution are illustrated with two known data sets from the literature: smoking cessation data and breast cosmesis data.

Estimation After a Group Sequential Trial.

PubMed

Milanzi, Elasma; Molenberghs, Geert; Alonso, Ariel; Kenward, Michael G; Tsiatis, Anastasios A; Davidian, Marie; Verbeke, Geert

2015-10-01

Group sequential trials are one important instance of studies for which the sample size is not fixed a priori but rather takes one of a finite set of pre-specified values, dependent on the observed data. Much work has been devoted to the inferential consequences of this design feature. Molenberghs et al (2012) and Milanzi et al (2012) reviewed and extended the existing literature, focusing on a collection of seemingly disparate, but related, settings, namely completely random sample sizes, group sequential studies with deterministic and random stopping rules, incomplete data, and random cluster sizes. They showed that the ordinary sample average is a viable option for estimation following a group sequential trial, for a wide class of stopping rules and for random outcomes with a distribution in the exponential family. Their results are somewhat surprising in the sense that the sample average is not optimal, and further, there does not exist an optimal, or even, unbiased linear estimator. However, the sample average is asymptotically unbiased, both conditionally upon the observed sample size as well as marginalized over it. By exploiting ignorability they showed that the sample average is the conventional maximum likelihood estimator. They also showed that a conditional maximum likelihood estimator is finite sample unbiased, but is less efficient than the sample average and has the larger mean squared error. Asymptotically, the sample average and the conditional maximum likelihood estimator are equivalent. This previous work is restricted, however, to the situation in which the the random sample size can take only two values, N = n or N = 2 n . In this paper, we consider the more practically useful setting of sample sizes in a the finite set { n 1 , n 2 , …, n L }. It is shown that the sample average is then a justifiable estimator , in the sense that it follows from joint likelihood estimation, and it is consistent and asymptotically unbiased. We also show why simulations can give the false impression of bias in the sample average when considered conditional upon the sample size. The consequence is that no corrections need to be made to estimators following sequential trials. When small-sample bias is of concern, the conditional likelihood estimator provides a relatively straightforward modification to the sample average. Finally, it is shown that classical likelihood-based standard errors and confidence intervals can be applied, obviating the need for technical corrections.

On the mass of the local group

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

González, Roberto E.; Kravtsov, Andrey V.; Gnedin, Nickolay Y., E-mail: regonzar@astro.puc.cl

2014-10-01

We use recent proper motion measurements of the tangential velocity of M31, along with its radial velocity and distance, to derive the likelihood of the sum of halo masses of the Milky Way and M31. This is done using a sample of halo pairs in the Bolshoi cosmological simulation of ΛCDM cosmology selected to match the properties and the environment of the Local Group. The resulting likelihood gives an estimate of the sum of the masses of M {sub MW,} {sub 200c} + M {sub M31,} {sub 200c} = 2.40{sub −1.05}{sup +1.95}×10{sup 12} M{sub ⊙} (90% confidence interval). This estimatemore » is consistent with individual mass estimates for the Milky Way and M31 and is consistent, albeit somewhat on the low side, with the mass estimated using the timing argument. We show that although the timing argument is unbiased on average for all pairs, for pairs constrained to have radial and tangential velocities similar to that of the Local Group the argument overestimates the sum of masses by a factor of 1.6. Using similar technique, we estimate the total dark matter mass enclosed within 1 Mpc from the Local Group barycenter to be M{sub LG}(r<1 Mpc)=4.2{sub −2.0}{sup +3.4}×10{sup 12} M{sub ⊙} (90% confidence interval).« less

Linkage disequilibrium interval mapping of quantitative trait loci.

PubMed

Boitard, Simon; Abdallah, Jihad; de Rochambeau, Hubert; Cierco-Ayrolles, Christine; Mangin, Brigitte

2006-03-16

For many years gene mapping studies have been performed through linkage analyses based on pedigree data. Recently, linkage disequilibrium methods based on unrelated individuals have been advocated as powerful tools to refine estimates of gene location. Many strategies have been proposed to deal with simply inherited disease traits. However, locating quantitative trait loci is statistically more challenging and considerable research is needed to provide robust and computationally efficient methods. Under a three-locus Wright-Fisher model, we derived approximate expressions for the expected haplotype frequencies in a population. We considered haplotypes comprising one trait locus and two flanking markers. Using these theoretical expressions, we built a likelihood-maximization method, called HAPim, for estimating the location of a quantitative trait locus. For each postulated position, the method only requires information from the two flanking markers. Over a wide range of simulation scenarios it was found to be more accurate than a two-marker composite likelihood method. It also performed as well as identity by descent methods, whilst being valuable in a wider range of populations. Our method makes efficient use of marker information, and can be valuable for fine mapping purposes. Its performance is increased if multiallelic markers are available. Several improvements can be developed to account for more complex evolution scenarios or provide robust confidence intervals for the location estimates.

An Algorithm for Efficient Maximum Likelihood Estimation and Confidence Interval Determination in Nonlinear Estimation Problems

NASA Technical Reports Server (NTRS)

Murphy, Patrick Charles

1985-01-01

An algorithm for maximum likelihood (ML) estimation is developed with an efficient method for approximating the sensitivities. The algorithm was developed for airplane parameter estimation problems but is well suited for most nonlinear, multivariable, dynamic systems. The ML algorithm relies on a new optimization method referred to as a modified Newton-Raphson with estimated sensitivities (MNRES). MNRES determines sensitivities by using slope information from local surface approximations of each output variable in parameter space. The fitted surface allows sensitivity information to be updated at each iteration with a significant reduction in computational effort. MNRES determines the sensitivities with less computational effort than using either a finite-difference method or integrating the analytically determined sensitivity equations. MNRES eliminates the need to derive sensitivity equations for each new model, thus eliminating algorithm reformulation with each new model and providing flexibility to use model equations in any format that is convenient. A random search technique for determining the confidence limits of ML parameter estimates is applied to nonlinear estimation problems for airplanes. The confidence intervals obtained by the search are compared with Cramer-Rao (CR) bounds at the same confidence level. It is observed that the degree of nonlinearity in the estimation problem is an important factor in the relationship between CR bounds and the error bounds determined by the search technique. The CR bounds were found to be close to the bounds determined by the search when the degree of nonlinearity was small. Beale's measure of nonlinearity is developed in this study for airplane identification problems; it is used to empirically correct confidence levels for the parameter confidence limits. The primary utility of the measure, however, was found to be in predicting the degree of agreement between Cramer-Rao bounds and search estimates.

Bayesian forecasting and uncertainty quantifying of stream flows using Metropolis–Hastings Markov Chain Monte Carlo algorithm

DOE PAGES

Wang, Hongrui; Wang, Cheng; Wang, Ying; ...

2017-04-05

This paper presents a Bayesian approach using Metropolis-Hastings Markov Chain Monte Carlo algorithm and applies this method for daily river flow rate forecast and uncertainty quantification for Zhujiachuan River using data collected from Qiaotoubao Gage Station and other 13 gage stations in Zhujiachuan watershed in China. The proposed method is also compared with the conventional maximum likelihood estimation (MLE) for parameter estimation and quantification of associated uncertainties. While the Bayesian method performs similarly in estimating the mean value of daily flow rate, it performs over the conventional MLE method on uncertainty quantification, providing relatively narrower reliable interval than the MLEmore » confidence interval and thus more precise estimation by using the related information from regional gage stations. As a result, the Bayesian MCMC method might be more favorable in the uncertainty analysis and risk management.« less

The Inverse Problem for Confined Aquifer Flow: Identification and Estimation With Extensions

NASA Astrophysics Data System (ADS)

Loaiciga, Hugo A.; MariñO, Miguel A.

1987-01-01

The contributions of this work are twofold. First, a methodology for estimating the elements of parameter matrices in the governing equation of flow in a confined aquifer is developed. The estimation techniques for the distributed-parameter inverse problem pertain to linear least squares and generalized least squares methods. The linear relationship among the known heads and unknown parameters of the flow equation provides the background for developing criteria for determining the identifiability status of unknown parameters. Under conditions of exact or overidentification it is possible to develop statistically consistent parameter estimators and their asymptotic distributions. The estimation techniques, namely, two-stage least squares and three stage least squares, are applied to a specific groundwater inverse problem and compared between themselves and with an ordinary least squares estimator. The three-stage estimator provides the closer approximation to the actual parameter values, but it also shows relatively large standard errors as compared to the ordinary and two-stage estimators. The estimation techniques provide the parameter matrices required to simulate the unsteady groundwater flow equation. Second, a nonlinear maximum likelihood estimation approach to the inverse problem is presented. The statistical properties of maximum likelihood estimators are derived, and a procedure to construct confidence intervals and do hypothesis testing is given. The relative merits of the linear and maximum likelihood estimators are analyzed. Other topics relevant to the identification and estimation methodologies, i.e., a continuous-time solution to the flow equation, coping with noise-corrupted head measurements, and extension of the developed theory to nonlinear cases are also discussed. A simulation study is used to evaluate the methods developed in this study.

Markov modulated Poisson process models incorporating covariates for rainfall intensity.

PubMed

Thayakaran, R; Ramesh, N I

2013-01-01

Time series of rainfall bucket tip times at the Beaufort Park station, Bracknell, in the UK are modelled by a class of Markov modulated Poisson processes (MMPP) which may be thought of as a generalization of the Poisson process. Our main focus in this paper is to investigate the effects of including covariate information into the MMPP model framework on statistical properties. In particular, we look at three types of time-varying covariates namely temperature, sea level pressure, and relative humidity that are thought to be affecting the rainfall arrival process. Maximum likelihood estimation is used to obtain the parameter estimates, and likelihood ratio tests are employed in model comparison. Simulated data from the fitted model are used to make statistical inferences about the accumulated rainfall in the discrete time interval. Variability of the daily Poisson arrival rates is studied.

Maximum likelihood estimation of correction for dilution bias in simple linear regression using replicates from subjects with extreme first measurements.

PubMed

Berglund, Lars; Garmo, Hans; Lindbäck, Johan; Svärdsudd, Kurt; Zethelius, Björn

2008-09-30

The least-squares estimator of the slope in a simple linear regression model is biased towards zero when the predictor is measured with random error. A corrected slope may be estimated by adding data from a reliability study, which comprises a subset of subjects from the main study. The precision of this corrected slope depends on the design of the reliability study and estimator choice. Previous work has assumed that the reliability study constitutes a random sample from the main study. A more efficient design is to use subjects with extreme values on their first measurement. Previously, we published a variance formula for the corrected slope, when the correction factor is the slope in the regression of the second measurement on the first. In this paper we show that both designs improve by maximum likelihood estimation (MLE). The precision gain is explained by the inclusion of data from all subjects for estimation of the predictor's variance and by the use of the second measurement for estimation of the covariance between response and predictor. The gain of MLE enhances with stronger true relationship between response and predictor and with lower precision in the predictor measurements. We present a real data example on the relationship between fasting insulin, a surrogate marker, and true insulin sensitivity measured by a gold-standard euglycaemic insulin clamp, and simulations, where the behavior of profile-likelihood-based confidence intervals is examined. MLE was shown to be a robust estimator for non-normal distributions and efficient for small sample situations. Copyright (c) 2008 John Wiley & Sons, Ltd.

Estimation of brood and nest survival: Comparative methods in the presence of heterogeneity

USGS Publications Warehouse

Manly, Bryan F.J.; Schmutz, Joel A.

2001-01-01

The Mayfield method has been widely used for estimating survival of nests and young animals, especially when data are collected at irregular observation intervals. However, this method assumes survival is constant throughout the study period, which often ignores biologically relevant variation and may lead to biased survival estimates. We examined the bias and accuracy of 1 modification to the Mayfield method that allows for temporal variation in survival, and we developed and similarly tested 2 additional methods. One of these 2 new methods is simply an iterative extension of Klett and Johnson's method, which we refer to as the Iterative Mayfield method and bears similarity to Kaplan-Meier methods. The other method uses maximum likelihood techniques for estimation and is best applied to survival of animals in groups or families, rather than as independent individuals. We also examined how robust these estimators are to heterogeneity in the data, which can arise from such sources as dependent survival probabilities among siblings, inherent differences among families, and adoption. Testing of estimator performance with respect to bias, accuracy, and heterogeneity was done using simulations that mimicked a study of survival of emperor goose (Chen canagica) goslings. Assuming constant survival for inappropriately long periods of time or use of Klett and Johnson's methods resulted in large bias or poor accuracy (often >5% bias or root mean square error) compared to our Iterative Mayfield or maximum likelihood methods. Overall, estimator performance was slightly better with our Iterative Mayfield than our maximum likelihood method, but the maximum likelihood method provides a more rigorous framework for testing covariates and explicity models a heterogeneity factor. We demonstrated use of all estimators with data from emperor goose goslings. We advocate that future studies use the new methods outlined here rather than the traditional Mayfield method or its previous modifications.

Estimation for coefficient of variation of an extension of the exponential distribution under type-II censoring scheme

NASA Astrophysics Data System (ADS)

Bakoban, Rana A.

2017-08-01

The coefficient of variation [CV] has several applications in applied statistics. So in this paper, we adopt Bayesian and non-Bayesian approaches for the estimation of CV under type-II censored data from extension exponential distribution [EED]. The point and interval estimate of the CV are obtained for each of the maximum likelihood and parametric bootstrap techniques. Also the Bayesian approach with the help of MCMC method is presented. A real data set is presented and analyzed, hence the obtained results are used to assess the obtained theoretical results.

An evaluation of inferential procedures for adaptive clinical trial designs with pre-specified rules for modifying the sample size.

PubMed

Levin, Gregory P; Emerson, Sarah C; Emerson, Scott S

2014-09-01

Many papers have introduced adaptive clinical trial methods that allow modifications to the sample size based on interim estimates of treatment effect. There has been extensive commentary on type I error control and efficiency considerations, but little research on estimation after an adaptive hypothesis test. We evaluate the reliability and precision of different inferential procedures in the presence of an adaptive design with pre-specified rules for modifying the sampling plan. We extend group sequential orderings of the outcome space based on the stage at stopping, likelihood ratio statistic, and sample mean to the adaptive setting in order to compute median-unbiased point estimates, exact confidence intervals, and P-values uniformly distributed under the null hypothesis. The likelihood ratio ordering is found to average shorter confidence intervals and produce higher probabilities of P-values below important thresholds than alternative approaches. The bias adjusted mean demonstrates the lowest mean squared error among candidate point estimates. A conditional error-based approach in the literature has the benefit of being the only method that accommodates unplanned adaptations. We compare the performance of this and other methods in order to quantify the cost of failing to plan ahead in settings where adaptations could realistically be pre-specified at the design stage. We find the cost to be meaningful for all designs and treatment effects considered, and to be substantial for designs frequently proposed in the literature. © 2014, The International Biometric Society.

New method to incorporate Type B uncertainty into least-squares procedures in radionuclide metrology.

PubMed

Han, Jubong; Lee, K B; Lee, Jong-Man; Park, Tae Soon; Oh, J S; Oh, Pil-Jei

2016-03-01

We discuss a new method to incorporate Type B uncertainty into least-squares procedures. The new method is based on an extension of the likelihood function from which a conventional least-squares function is derived. The extended likelihood function is the product of the original likelihood function with additional PDFs (Probability Density Functions) that characterize the Type B uncertainties. The PDFs are considered to describe one's incomplete knowledge on correction factors being called nuisance parameters. We use the extended likelihood function to make point and interval estimations of parameters in the basically same way as the least-squares function used in the conventional least-squares method is derived. Since the nuisance parameters are not of interest and should be prevented from appearing in the final result, we eliminate such nuisance parameters by using the profile likelihood. As an example, we present a case study for a linear regression analysis with a common component of Type B uncertainty. In this example we compare the analysis results obtained from using our procedure with those from conventional methods. Copyright © 2015. Published by Elsevier Ltd.

Cox model with interval-censored covariate in cohort studies.

PubMed

Ahn, Soohyun; Lim, Johan; Paik, Myunghee Cho; Sacco, Ralph L; Elkind, Mitchell S

2018-05-18

In cohort studies the outcome is often time to a particular event, and subjects are followed at regular intervals. Periodic visits may also monitor a secondary irreversible event influencing the event of primary interest, and a significant proportion of subjects develop the secondary event over the period of follow-up. The status of the secondary event serves as a time-varying covariate, but is recorded only at the times of the scheduled visits, generating incomplete time-varying covariates. While information on a typical time-varying covariate is missing for entire follow-up period except the visiting times, the status of the secondary event are unavailable only between visits where the status has changed, thus interval-censored. One may view interval-censored covariate of the secondary event status as missing time-varying covariates, yet missingness is partial since partial information is provided throughout the follow-up period. Current practice of using the latest observed status produces biased estimators, and the existing missing covariate techniques cannot accommodate the special feature of missingness due to interval censoring. To handle interval-censored covariates in the Cox proportional hazards model, we propose an available-data estimator, a doubly robust-type estimator as well as the maximum likelihood estimator via EM algorithm and present their asymptotic properties. We also present practical approaches that are valid. We demonstrate the proposed methods using our motivating example from the Northern Manhattan Study. © 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Likelihood of home death associated with local rates of home birth: influence of local area healthcare preferences on site of death.

PubMed

Silveira, Maria J; Copeland, Laurel A; Feudtner, Chris

2006-07-01

We tested whether local cultural and social values regarding the use of health care are associated with the likelihood of home death, using variation in local rates of home births as a proxy for geographic variation in these values. For each of 351110 adult decedents in Washington state who died from 1989 through 1998, we calculated the home birth rate in each zip code during the year of death and then used multivariate regression modeling to estimate the relation between the likelihood of home death and the local rate of home births. Individuals residing in local areas with higher home birth rates had greater adjusted likelihood of dying at home (odds ratio [OR]=1.04 for each percentage point increase in home birth rate; 95% confidence interval [CI] = 1.03, 1.05). Moreover, the likelihood of dying at home increased with local wealth (OR=1.04 per $10000; 95% CI=1.02, 1.06) but decreased with local hospital bed availability (OR=0.96 per 1000 beds; 95% CI=0.95, 0.97). The likelihood of home death is associated with local rates of home births, suggesting the influence of health care use preferences.

Modeling the distribution of extreme share return in Malaysia using Generalized Extreme Value (GEV) distribution

NASA Astrophysics Data System (ADS)

Hasan, Husna; Radi, Noor Fadhilah Ahmad; Kassim, Suraiya

2012-05-01

Extreme share return in Malaysia is studied. The monthly, quarterly, half yearly and yearly maximum returns are fitted to the Generalized Extreme Value (GEV) distribution. The Augmented Dickey Fuller (ADF) and Phillips Perron (PP) tests are performed to test for stationarity, while Mann-Kendall (MK) test is for the presence of monotonic trend. Maximum Likelihood Estimation (MLE) is used to estimate the parameter while L-moments estimate (LMOM) is used to initialize the MLE optimization routine for the stationary model. Likelihood ratio test is performed to determine the best model. Sherman's goodness of fit test is used to assess the quality of convergence of the GEV distribution by these monthly, quarterly, half yearly and yearly maximum. Returns levels are then estimated for prediction and planning purposes. The results show all maximum returns for all selection periods are stationary. The Mann-Kendall test indicates the existence of trend. Thus, we ought to model for non-stationary model too. Model 2, where the location parameter is increasing with time is the best for all selection intervals. Sherman's goodness of fit test shows that monthly, quarterly, half yearly and yearly maximum converge to the GEV distribution. From the results, it seems reasonable to conclude that yearly maximum is better for the convergence to the GEV distribution especially if longer records are available. Return level estimates, which is the return level (in this study return amount) that is expected to be exceeded, an average, once every t time periods starts to appear in the confidence interval of T = 50 for quarterly, half yearly and yearly maximum.

Accounting for dropout bias using mixed-effects models.

PubMed

Mallinckrodt, C H; Clark, W S; David, S R

2001-01-01

Treatment effects are often evaluated by comparing change over time in outcome measures. However, valid analyses of longitudinal data can be problematic when subjects discontinue (dropout) prior to completing the study. This study assessed the merits of likelihood-based repeated measures analyses (MMRM) compared with fixed-effects analysis of variance where missing values were imputed using the last observation carried forward approach (LOCF) in accounting for dropout bias. Comparisons were made in simulated data and in data from a randomized clinical trial. Subject dropout was introduced in the simulated data to generate ignorable and nonignorable missingness. Estimates of treatment group differences in mean change from baseline to endpoint from MMRM were, on average, markedly closer to the true value than estimates from LOCF in every scenario simulated. Standard errors and confidence intervals from MMRM accurately reflected the uncertainty of the estimates, whereas standard errors and confidence intervals from LOCF underestimated uncertainty.

Estimating ambiguity preferences and perceptions in multiple prior models: Evidence from the field.

PubMed

Dimmock, Stephen G; Kouwenberg, Roy; Mitchell, Olivia S; Peijnenburg, Kim

2015-12-01

We develop a tractable method to estimate multiple prior models of decision-making under ambiguity. In a representative sample of the U.S. population, we measure ambiguity attitudes in the gain and loss domains. We find that ambiguity aversion is common for uncertain events of moderate to high likelihood involving gains, but ambiguity seeking prevails for low likelihoods and for losses. We show that choices made under ambiguity in the gain domain are best explained by the α-MaxMin model, with one parameter measuring ambiguity aversion (ambiguity preferences) and a second parameter quantifying the perceived degree of ambiguity (perceptions about ambiguity). The ambiguity aversion parameter α is constant and prior probability sets are asymmetric for low and high likelihood events. The data reject several other models, such as MaxMin and MaxMax, as well as symmetric probability intervals. Ambiguity aversion and the perceived degree of ambiguity are both higher for men and for the college-educated. Ambiguity aversion (but not perceived ambiguity) is also positively related to risk aversion. In the loss domain, we find evidence of reflection, implying that ambiguity aversion for gains tends to reverse into ambiguity seeking for losses. Our model's estimates for preferences and perceptions about ambiguity can be used to analyze the economic and financial implications of such preferences.

Population density estimated from locations of individuals on a passive detector array

USGS Publications Warehouse

Efford, Murray G.; Dawson, Deanna K.; Borchers, David L.

2009-01-01

The density of a closed population of animals occupying stable home ranges may be estimated from detections of individuals on an array of detectors, using newly developed methods for spatially explicit capture–recapture. Likelihood-based methods provide estimates for data from multi-catch traps or from devices that record presence without restricting animal movement ("proximity" detectors such as camera traps and hair snags). As originally proposed, these methods require multiple sampling intervals. We show that equally precise and unbiased estimates may be obtained from a single sampling interval, using only the spatial pattern of detections. This considerably extends the range of possible applications, and we illustrate the potential by estimating density from simulated detections of bird vocalizations on a microphone array. Acoustic detection can be defined as occurring when received signal strength exceeds a threshold. We suggest detection models for binary acoustic data, and for continuous data comprising measurements of all signals above the threshold. While binary data are often sufficient for density estimation, modeling signal strength improves precision when the microphone array is small.

Estimating degradation in real time and accelerated stability tests with random lot-to-lot variation: a simulation study.

PubMed

Magari, Robert T

2002-03-01

The effect of different lot-to-lot variability levels on the prediction of stability are studied based on two statistical models for estimating degradation in real time and accelerated stability tests. Lot-to-lot variability is considered as random in both models, and is attributed to two sources-variability at time zero, and variability of degradation rate. Real-time stability tests are modeled as a function of time while accelerated stability tests as a function of time and temperatures. Several data sets were simulated, and a maximum likelihood approach was used for estimation. The 95% confidence intervals for the degradation rate depend on the amount of lot-to-lot variability. When lot-to-lot degradation rate variability is relatively large (CV > or = 8%) the estimated confidence intervals do not represent the trend for individual lots. In such cases it is recommended to analyze each lot individually. Copyright 2002 Wiley-Liss, Inc. and the American Pharmaceutical Association J Pharm Sci 91: 893-899, 2002

Generalized likelihood ratios for quantitative diagnostic test scores.

PubMed

Tandberg, D; Deely, J J; O'Malley, A J

1997-11-01

The reduction of quantitative diagnostic test scores to the dichotomous case is a wasteful and unnecessary simplification in the era of high-speed computing. Physicians could make better use of the information embedded in quantitative test results if modern generalized curve estimation techniques were applied to the likelihood functions of Bayes' theorem. Hand calculations could be completely avoided and computed graphical summaries provided instead. Graphs showing posttest probability of disease as a function of pretest probability with confidence intervals (POD plots) would enhance acceptance of these techniques if they were immediately available at the computer terminal when test results were retrieved. Such constructs would also provide immediate feedback to physicians when a valueless test had been ordered.

Empirical likelihood-based confidence intervals for mean medical cost with censored data.

PubMed

Jeyarajah, Jenny; Qin, Gengsheng

2017-11-10

In this paper, we propose empirical likelihood methods based on influence function and jackknife techniques for constructing confidence intervals for mean medical cost with censored data. We conduct a simulation study to compare the coverage probabilities and interval lengths of our proposed confidence intervals with that of the existing normal approximation-based confidence intervals and bootstrap confidence intervals. The proposed methods have better finite-sample performances than existing methods. Finally, we illustrate our proposed methods with a relevant example. Copyright © 2017 John Wiley & Sons, Ltd.

Probabilities and statistics for backscatter estimates obtained by a scatterometer with applications to new scatterometer design data

NASA Technical Reports Server (NTRS)

Pierson, Willard J., Jr.

1989-01-01

The values of the Normalized Radar Backscattering Cross Section (NRCS), sigma (o), obtained by a scatterometer are random variables whose variance is a known function of the expected value. The probability density function can be obtained from the normal distribution. Models for the expected value obtain it as a function of the properties of the waves on the ocean and the winds that generated the waves. Point estimates of the expected value were found from various statistics given the parameters that define the probability density function for each value. Random intervals were derived with a preassigned probability of containing that value. A statistical test to determine whether or not successive values of sigma (o) are truly independent was derived. The maximum likelihood estimates for wind speed and direction were found, given a model for backscatter as a function of the properties of the waves on the ocean. These estimates are biased as a result of the terms in the equation that involve natural logarithms, and calculations of the point estimates of the maximum likelihood values are used to show that the contributions of the logarithmic terms are negligible and that the terms can be omitted.

Statistical inferences with jointly type-II censored samples from two Pareto distributions

NASA Astrophysics Data System (ADS)

Abu-Zinadah, Hanaa H.

2017-08-01

In the several fields of industries the product comes from more than one production line, which is required to work the comparative life tests. This problem requires sampling of the different production lines, then the joint censoring scheme is appeared. In this article we consider the life time Pareto distribution with jointly type-II censoring scheme. The maximum likelihood estimators (MLE) and the corresponding approximate confidence intervals as well as the bootstrap confidence intervals of the model parameters are obtained. Also Bayesian point and credible intervals of the model parameters are presented. The life time data set is analyzed for illustrative purposes. Monte Carlo results from simulation studies are presented to assess the performance of our proposed method.

Minimizing pre- and post-defibrillation pauses increases the likelihood of return of spontaneous circulation (ROSC).

PubMed

Sell, Rebecca E; Sarno, Renee; Lawrence, Brenna; Castillo, Edward M; Fisher, Roger; Brainard, Criss; Dunford, James V; Davis, Daniel P

2010-07-01

The three-phase model of ventricular fibrillation (VF) arrest suggests a period of compressions to "prime" the heart prior to defibrillation attempts. In addition, post-shock compressions may increase the likelihood of return of spontaneous circulation (ROSC). The optimal intervals for shock delivery following cessation of compressions (pre-shock interval) and resumption of compressions following a shock (post-shock interval) remain unclear. To define optimal pre- and post-defibrillation compression pauses for out-of-hospital cardiac arrest (OOHCA). All patients suffering OOHCA from VF were identified over a 1-month period. Defibrillator data were abstracted and analyzed using the combination of ECG, impedance, and audio recording. Receiver-operator curve (ROC) analysis was used to define the optimal pre- and post-shock compression intervals. Multiple logistic regression analysis was used to quantify the relationship between these intervals and ROSC. Covariates included cumulative number of defibrillation attempts, intubation status, and administration of epinephrine in the immediate pre-shock compression cycle. Cluster adjustment was performed due to the possibility of multiple defibrillation attempts for each patient. A total of 36 patients with 96 defibrillation attempts were included. The ROC analysis identified an optimal pre-shock interval of <3s and an optimal post-shock interval of <6s. Increased likelihood of ROSC was observed with a pre-shock interval <3s (adjusted OR 6.7, 95% CI 2.0-22.3, p=0.002) and a post-shock interval of <6s (adjusted OR 10.7, 95% CI 2.8-41.4, p=0.001). Likelihood of ROSC was substantially increased with the optimization of both pre- and post-shock intervals (adjusted OR 13.1, 95% CI 3.4-49.9, p<0.001). Decreasing pre- and post-shock compression intervals increases the likelihood of ROSC in OOHCA from VF.

Estimation of Gutenberg-Richter b-value using instrumental earthquake catalog from the southern Korean Peninsula

NASA Astrophysics Data System (ADS)

Lee, H.; Sheen, D.; Kim, S.

2013-12-01

The b-value in Gutenberg-Richter relation is an important parameter widely used not only in the interpretation of regional tectonic structure but in the seismic hazard analysis. In this study, we tested four methods for estimating the stable b-value in a small number of events using Monte-Carlo method. One is the Least-Squares method (LSM) which minimizes the observation error. Others are based on the Maximum Likelihood method (MLM) which maximizes the likelihood function: Utsu's (1965) method for continuous magnitudes and an infinite maximum magnitude, Page's (1968) for continuous magnitudes and a finite maximum magnitude, and Weichert's (1980) for interval magnitude and a finite maximum magnitude. A synthetic parent population of the earthquake catalog of million events from magnitude 2.0 to 7.0 with interval of 0.1 was generated for the Monte-Carlo simulation. The sample, the number of which was increased from 25 to 1000, was extracted from the parent population randomly. The resampling procedure was applied 1000 times with different random seed numbers. The mean and the standard deviation of the b-value were estimated for each sample group that has the same number of samples. As expected, the more samples were used, the more stable b-value was obtained. However, in a small number of events, the LSM gave generally low b-value with a large standard deviation while other MLMs gave more accurate and stable values. It was found that Utsu (1965) gives the most accurate and stable b-value even in a small number of events. It was also found that the selection of the minimum magnitude could be critical for estimating the correct b-value for Utsu's (1965) method and Page's (1968) if magnitudes were binned into an interval. Therefore, we applied Utsu (1965) to estimate the b-value using two instrumental earthquake catalogs, which have events occurred around the southern part of the Korean Peninsula from 1978 to 2011. By a careful choice of the minimum magnitude, the b-values of the earthquake catalogs of the Korea Meteorological Administration and Kim (2012) are estimated to be 0.72 and 0.74, respectively.

Can 3-dimensional power Doppler indices improve the prenatal diagnosis of a potentially morbidly adherent placenta in patients with placenta previa?

PubMed

Haidar, Ziad A; Papanna, Ramesha; Sibai, Baha M; Tatevian, Nina; Viteri, Oscar A; Vowels, Patricia C; Blackwell, Sean C; Moise, Kenneth J

2017-08-01

Traditionally, 2-dimensional ultrasound parameters have been used for the diagnosis of a suspected morbidly adherent placenta previa. More objective techniques have not been well studied yet. The objective of the study was to determine the ability of prenatal 3-dimensional power Doppler analysis of flow and vascular indices to predict the morbidly adherent placenta objectively. A prospective cohort study was performed in women between 28 and 32 gestational weeks with known placenta previa. Patients underwent a two-dimensional gray-scale ultrasound that determined management decisions. 3-Dimensional power Doppler volumes were obtained during the same examination and vascular, flow, and vascular flow indices were calculated after manual tracing of the viewed placenta in the sweep; data were blinded to obstetricians. Morbidly adherent placenta was confirmed by histology. Severe morbidly adherent placenta was defined as increta/percreta on histology, blood loss >2000 mL, and >2 units of PRBC transfused. Sensitivities, specificities, predictive values, and likelihood ratios were calculated. Student t and χ 2 tests, logistic regression, receiver-operating characteristic curves, and intra- and interrater agreements using Kappa statistics were performed. The following results were found: (1) 50 women were studied: 23 had morbidly adherent placenta, of which 12 (52.2%) were severe morbidly adherent placenta; (2) 2-dimensional parameters diagnosed morbidly adherent placenta with a sensitivity of 82.6% (95% confidence interval, 60.4-94.2), a specificity of 88.9% (95% confidence interval, 69.7-97.1), a positive predictive value of 86.3% (95% confidence interval, 64.0-96.4), a negative predictive value of 85.7% (95% confidence interval, 66.4-95.3), a positive likelihood ratio of 7.4 (95% confidence interval, 2.5-21.9), and a negative likelihood ratio of 0.2 (95% confidence interval, 0.08-0.48); (3) mean values of the vascular index (32.8 ± 7.4) and the vascular flow index (14.2 ± 3.8) were higher in morbidly adherent placenta (P < .001); (4) area under the receiver-operating characteristic curve for the vascular and vascular flow indices were 0.99 and 0.97, respectively; (5) the vascular index ≥21 predicted morbidly adherent placenta with a sensitivity and a specificity of 95% (95% confidence interval, 88.2-96.9) and 91%, respectively (95% confidence interval, 87.5-92.4), 92% positive predictive value (95% confidence interval, 85.5-94.3), 90% negative predictive value (95% confidence interval, 79.9-95.3), positive likelihood ratio of 10.55 (95% confidence interval, 7.06-12.75), and negative likelihood ratio of 0.05 (95% confidence interval, 0.03-0.13); and (6) for the severe morbidly adherent placenta, 2-dimensional ultrasound had a sensitivity of 33.3% (95% confidence interval, 11.3-64.6), a specificity of 81.8% (95% confidence interval, 47.8-96.8), a positive predictive value of 66.7% (95% confidence interval, 24.1-94.1), a negative predictive value of 52.9% (95% confidence interval, 28.5-76.1), a positive likelihood ratio of 1.83 (95% confidence interval, 0.41-8.11), and a negative likelihood ratio of 0.81 (95% confidence interval, 0.52-1.26). A vascular index ≥31 predicted the diagnosis of a severe morbidly adherent placenta with a 100% sensitivity (95% confidence interval, 72-100), a 90% specificity (95% confidence interval, 81.7-93.8), an 88% positive predictive value (95% confidence interval, 55.0-91.3), a 100% negative predictive value (95% confidence interval, 90.9-100), a positive likelihood ratio of 10.0 (95% confidence interval, 3.93-16.13), and a negative likelihood ratio of 0 (95% confidence interval, 0-0.34). Intrarater and interrater agreements were 94% (P < .001) and 93% (P < .001), respectively. The vascular index accurately predicts the morbidly adherent placenta in patients with placenta previa. In addition, 3-dimensional power Doppler vascular and vascular flow indices were more predictive of severe cases of morbidly adherent placenta compared with 2-dimensional ultrasound. This objective technique may limit the variations in diagnosing morbidly adherent placenta because of the subjectivity of 2-dimensional ultrasound interpretations. Copyright © 2017 Elsevier Inc. All rights reserved.

Uncertainty in Population Growth Rates: Determining Confidence Intervals from Point Estimates of Parameters

PubMed Central

Devenish Nelson, Eleanor S.; Harris, Stephen; Soulsbury, Carl D.; Richards, Shane A.; Stephens, Philip A.

2010-01-01

Background Demographic models are widely used in conservation and management, and their parameterisation often relies on data collected for other purposes. When underlying data lack clear indications of associated uncertainty, modellers often fail to account for that uncertainty in model outputs, such as estimates of population growth. Methodology/Principal Findings We applied a likelihood approach to infer uncertainty retrospectively from point estimates of vital rates. Combining this with resampling techniques and projection modelling, we show that confidence intervals for population growth estimates are easy to derive. We used similar techniques to examine the effects of sample size on uncertainty. Our approach is illustrated using data on the red fox, Vulpes vulpes, a predator of ecological and cultural importance, and the most widespread extant terrestrial mammal. We show that uncertainty surrounding estimated population growth rates can be high, even for relatively well-studied populations. Halving that uncertainty typically requires a quadrupling of sampling effort. Conclusions/Significance Our results compel caution when comparing demographic trends between populations without accounting for uncertainty. Our methods will be widely applicable to demographic studies of many species. PMID:21049049

Halo-independent determination of the unmodulated WIMP signal in DAMA: the isotropic case

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

Gondolo, Paolo; Scopel, Stefano, E-mail: paolo.gondolo@utah.edu, E-mail: scopel@sogang.ac.kr

2017-09-01

We present a halo-independent determination of the unmodulated signal corresponding to the DAMA modulation if interpreted as due to dark matter weakly interacting massive particles (WIMPs). First we show how a modulated signal gives information on the WIMP velocity distribution function in the Galactic rest frame from which the unmodulated signal descends. Then we describe a mathematically-sound profile likelihood analysis in which the likelihood is profiled over a continuum of nuisance parameters (namely, the WIMP velocity distribution). As a first application of the method, which is very general and valid for any class of velocity distributions, we restrict the analysismore » to velocity distributions that are isotropic in the Galactic frame. In this way we obtain halo-independent maximum-likelihood estimates and confidence intervals for the DAMA unmodulated signal. We find that the estimated unmodulated signal is in line with expectations for a WIMP-induced modulation and is compatible with the DAMA background+signal rate. Specifically, for the isotropic case we find that the modulated amplitude ranges between a few percent and about 25% of the unmodulated amplitude, depending on the WIMP mass.« less

Parameter estimation of history-dependent leaky integrate-and-fire neurons using maximum-likelihood methods

PubMed Central

Dong, Yi; Mihalas, Stefan; Russell, Alexander; Etienne-Cummings, Ralph; Niebur, Ernst

2012-01-01

When a neuronal spike train is observed, what can we say about the properties of the neuron that generated it? A natural way to answer this question is to make an assumption about the type of neuron, select an appropriate model for this type, and then to choose the model parameters as those that are most likely to generate the observed spike train. This is the maximum likelihood method. If the neuron obeys simple integrate and fire dynamics, Paninski, Pillow, and Simoncelli (2004) showed that its negative log-likelihood function is convex and that its unique global minimum can thus be found by gradient descent techniques. The global minimum property requires independence of spike time intervals. Lack of history dependence is, however, an important constraint that is not fulfilled in many biological neurons which are known to generate a rich repertoire of spiking behaviors that are incompatible with history independence. Therefore, we expanded the integrate and fire model by including one additional variable, a variable threshold (Mihalas & Niebur, 2009) allowing for history-dependent firing patterns. This neuronal model produces a large number of spiking behaviors while still being linear. Linearity is important as it maintains the distribution of the random variables and still allows for maximum likelihood methods to be used. In this study we show that, although convexity of the negative log-likelihood is not guaranteed for this model, the minimum of the negative log-likelihood function yields a good estimate for the model parameters, in particular if the noise level is treated as a free parameter. Furthermore, we show that a nonlinear function minimization method (r-algorithm with space dilation) frequently reaches the global minimum. PMID:21851282

Relationship between two indicators of coronary risk estimated by the Framingham Risk Score and the number of metabolic syndrome components in Japanese male manufacturing workers.

PubMed

Kawada, Tomoyuki; Otsuka, Toshiaki; Inagaki, Hirofumi; Wakayama, Yoko; Li, Qing; Katsumata, Masao

2009-10-01

The Framingham Risk Score (FRS) has frequently been used in the United States to predict the 10-year risk of coronary heart disease (CHD). Components of the metabolic syndrome and several lifestyle factors have also been evaluated to estimate the risk of CHD. To determine the relationship between the FRS and components of metabolic syndrome as coronary risk indicators, the authors conducted a cross-sectional study of 2,619 Japanese male workers, ranging in age from 40 to 64 years, at a single workplace. Although the estimation by the FRS and metabolic syndrome involved some different factors, significant association of the risk estimated by the 2 methods was observed. When logistic regression analysis was conducted with adjustment for several lifestyle factors, the FRS and serum insulin were found to be significantly associated with the risk of likelihood of metabolic syndrome. The odds ratios and 95% confidence intervals of FRS by per standard deviation increment and serum insulin by increasing 1 microIU/mL for the prediction of metabolic syndrome were 2.50 (2.17-2.88) and 1.24 (1.20-1.27), respectively. A preventive effect of abstaining from drinking every day and eating breakfast almost daily against the likelihood of metabolic syndrome was also observed. In conclusion, the FRS and insulin were found to be significantly associated with the risk of likelihood of metabolic syndrome, even after controlling for weight change.

Interval estimation and optimal design for the within-subject coefficient of variation for continuous and binary variables

PubMed Central

Shoukri, Mohamed M; Elkum, Nasser; Walter, Stephen D

2006-01-01

Background In this paper we propose the use of the within-subject coefficient of variation as an index of a measurement's reliability. For continuous variables and based on its maximum likelihood estimation we derive a variance-stabilizing transformation and discuss confidence interval construction within the framework of a one-way random effects model. We investigate sample size requirements for the within-subject coefficient of variation for continuous and binary variables. Methods We investigate the validity of the approximate normal confidence interval by Monte Carlo simulations. In designing a reliability study, a crucial issue is the balance between the number of subjects to be recruited and the number of repeated measurements per subject. We discuss efficiency of estimation and cost considerations for the optimal allocation of the sample resources. The approach is illustrated by an example on Magnetic Resonance Imaging (MRI). We also discuss the issue of sample size estimation for dichotomous responses with two examples. Results For the continuous variable we found that the variance stabilizing transformation improves the asymptotic coverage probabilities on the within-subject coefficient of variation for the continuous variable. The maximum like estimation and sample size estimation based on pre-specified width of confidence interval are novel contribution to the literature for the binary variable. Conclusion Using the sample size formulas, we hope to help clinical epidemiologists and practicing statisticians to efficiently design reliability studies using the within-subject coefficient of variation, whether the variable of interest is continuous or binary. PMID:16686943

Selected aspects of prior and likelihood information for a Bayesian classifier in a road safety analysis.

PubMed

Nowakowska, Marzena

2017-04-01

The development of the Bayesian logistic regression model classifying the road accident severity is discussed. The already exploited informative priors (method of moments, maximum likelihood estimation, and two-stage Bayesian updating), along with the original idea of a Boot prior proposal, are investigated when no expert opinion has been available. In addition, two possible approaches to updating the priors, in the form of unbalanced and balanced training data sets, are presented. The obtained logistic Bayesian models are assessed on the basis of a deviance information criterion (DIC), highest probability density (HPD) intervals, and coefficients of variation estimated for the model parameters. The verification of the model accuracy has been based on sensitivity, specificity and the harmonic mean of sensitivity and specificity, all calculated from a test data set. The models obtained from the balanced training data set have a better classification quality than the ones obtained from the unbalanced training data set. The two-stage Bayesian updating prior model and the Boot prior model, both identified with the use of the balanced training data set, outperform the non-informative, method of moments, and maximum likelihood estimation prior models. It is important to note that one should be careful when interpreting the parameters since different priors can lead to different models. Copyright © 2017 Elsevier Ltd. All rights reserved.

Estimating ambiguity preferences and perceptions in multiple prior models: Evidence from the field

PubMed Central

Dimmock, Stephen G.; Kouwenberg, Roy; Mitchell, Olivia S.; Peijnenburg, Kim

2016-01-01

We develop a tractable method to estimate multiple prior models of decision-making under ambiguity. In a representative sample of the U.S. population, we measure ambiguity attitudes in the gain and loss domains. We find that ambiguity aversion is common for uncertain events of moderate to high likelihood involving gains, but ambiguity seeking prevails for low likelihoods and for losses. We show that choices made under ambiguity in the gain domain are best explained by the α-MaxMin model, with one parameter measuring ambiguity aversion (ambiguity preferences) and a second parameter quantifying the perceived degree of ambiguity (perceptions about ambiguity). The ambiguity aversion parameter α is constant and prior probability sets are asymmetric for low and high likelihood events. The data reject several other models, such as MaxMin and MaxMax, as well as symmetric probability intervals. Ambiguity aversion and the perceived degree of ambiguity are both higher for men and for the college-educated. Ambiguity aversion (but not perceived ambiguity) is also positively related to risk aversion. In the loss domain, we find evidence of reflection, implying that ambiguity aversion for gains tends to reverse into ambiguity seeking for losses. Our model’s estimates for preferences and perceptions about ambiguity can be used to analyze the economic and financial implications of such preferences. PMID:26924890

Modeling and E-M estimation of haplotype-specific relative risks from genotype data for a case-control study of unrelated individuals.

PubMed

Stram, Daniel O; Leigh Pearce, Celeste; Bretsky, Phillip; Freedman, Matthew; Hirschhorn, Joel N; Altshuler, David; Kolonel, Laurence N; Henderson, Brian E; Thomas, Duncan C

2003-01-01

The US National Cancer Institute has recently sponsored the formation of a Cohort Consortium (http://2002.cancer.gov/scpgenes.htm) to facilitate the pooling of data on very large numbers of people, concerning the effects of genes and environment on cancer incidence. One likely goal of these efforts will be generate a large population-based case-control series for which a number of candidate genes will be investigated using SNP haplotype as well as genotype analysis. The goal of this paper is to outline the issues involved in choosing a method of estimating haplotype-specific risk estimates for such data that is technically appropriate and yet attractive to epidemiologists who are already comfortable with odds ratios and logistic regression. Our interest is to develop and evaluate extensions of methods, based on haplotype imputation, that have been recently described (Schaid et al., Am J Hum Genet, 2002, and Zaykin et al., Hum Hered, 2002) as providing score tests of the null hypothesis of no effect of SNP haplotypes upon risk, which may be used for more complex tasks, such as providing confidence intervals, and tests of equivalence of haplotype-specific risks in two or more separate populations. In order to do so we (1) develop a cohort approach towards odds ratio analysis by expanding the E-M algorithm to provide maximum likelihood estimates of haplotype-specific odds ratios as well as genotype frequencies; (2) show how to correct the cohort approach, to give essentially unbiased estimates for population-based or nested case-control studies by incorporating the probability of selection as a case or control into the likelihood, based on a simplified model of case and control selection, and (3) finally, in an example data set (CYP17 and breast cancer, from the Multiethnic Cohort Study) we compare likelihood-based confidence interval estimates from the two methods with each other, and with the use of the single-imputation approach of Zaykin et al. applied under both null and alternative hypotheses. We conclude that so long as haplotypes are well predicted by SNP genotypes (we use the Rh2 criteria of Stram et al. [1]) the differences between the three methods are very small and in particular that the single imputation method may be expected to work extremely well. Copyright 2003 S. Karger AG, Basel

Estimating a Markovian Epidemic Model Using Household Serial Interval Data from the Early Phase of an Epidemic

PubMed Central

Black, Andrew J.; Ross, Joshua V.

2013-01-01

The clinical serial interval of an infectious disease is the time between date of symptom onset in an index case and the date of symptom onset in one of its secondary cases. It is a quantity which is commonly collected during a pandemic and is of fundamental importance to public health policy and mathematical modelling. In this paper we present a novel method for calculating the serial interval distribution for a Markovian model of household transmission dynamics. This allows the use of Bayesian MCMC methods, with explicit evaluation of the likelihood, to fit to serial interval data and infer parameters of the underlying model. We use simulated and real data to verify the accuracy of our methodology and illustrate the importance of accounting for household size. The output of our approach can be used to produce posterior distributions of population level epidemic characteristics. PMID:24023679

Application of change-point problem to the detection of plant patches.

PubMed

López, I; Gámez, M; Garay, J; Standovár, T; Varga, Z

2010-03-01

In ecology, if the considered area or space is large, the spatial distribution of individuals of a given plant species is never homogeneous; plants form different patches. The homogeneity change in space or in time (in particular, the related change-point problem) is an important research subject in mathematical statistics. In the paper, for a given data system along a straight line, two areas are considered, where the data of each area come from different discrete distributions, with unknown parameters. In the paper a method is presented for the estimation of the distribution change-point between both areas and an estimate is given for the distributions separated by the obtained change-point. The solution of this problem will be based on the maximum likelihood method. Furthermore, based on an adaptation of the well-known bootstrap resampling, a method for the estimation of the so-called change-interval is also given. The latter approach is very general, since it not only applies in the case of the maximum-likelihood estimation of the change-point, but it can be also used starting from any other change-point estimation known in the ecological literature. The proposed model is validated against typical ecological situations, providing at the same time a verification of the applied algorithms.

Calculation of Weibull strength parameters and Batdorf flow-density constants for volume- and surface-flaw-induced fracture in ceramics

NASA Technical Reports Server (NTRS)

Shantaram, S. Pai; Gyekenyesi, John P.

1989-01-01

The calculation of shape and scale parametes of the two-parameter Weibull distribution is described using the least-squares analysis and maximum likelihood methods for volume- and surface-flaw-induced fracture in ceramics with complete and censored samples. Detailed procedures are given for evaluating 90 percent confidence intervals for maximum likelihood estimates of shape and scale parameters, the unbiased estimates of the shape parameters, and the Weibull mean values and corresponding standard deviations. Furthermore, the necessary steps are described for detecting outliers and for calculating the Kolmogorov-Smirnov and the Anderson-Darling goodness-of-fit statistics and 90 percent confidence bands about the Weibull distribution. It also shows how to calculate the Batdorf flaw-density constants by using the Weibull distribution statistical parameters. The techniques described were verified with several example problems, from the open literature, and were coded in the Structural Ceramics Analysis and Reliability Evaluation (SCARE) design program.

Empirical likelihood-based confidence intervals for the sensitivity of a continuous-scale diagnostic test at a fixed level of specificity.

PubMed

Gengsheng Qin; Davis, Angela E; Jing, Bing-Yi

2011-06-01

For a continuous-scale diagnostic test, it is often of interest to find the range of the sensitivity of the test at the cut-off that yields a desired specificity. In this article, we first define a profile empirical likelihood ratio for the sensitivity of a continuous-scale diagnostic test and show that its limiting distribution is a scaled chi-square distribution. We then propose two new empirical likelihood-based confidence intervals for the sensitivity of the test at a fixed level of specificity by using the scaled chi-square distribution. Simulation studies are conducted to compare the finite sample performance of the newly proposed intervals with the existing intervals for the sensitivity in terms of coverage probability. A real example is used to illustrate the application of the recommended methods.

Reanalysis of cancer mortality in Japanese A-bomb survivors exposed to low doses of radiation: bootstrap and simulation methods

PubMed Central

2009-01-01

Background The International Commission on Radiological Protection (ICRP) recommended annual occupational dose limit is 20 mSv. Cancer mortality in Japanese A-bomb survivors exposed to less than 20 mSv external radiation in 1945 was analysed previously, using a latency model with non-linear dose response. Questions were raised regarding statistical inference with this model. Methods Cancers with over 100 deaths in the 0 - 20 mSv subcohort of the 1950-1990 Life Span Study are analysed with Poisson regression models incorporating latency, allowing linear and non-linear dose response. Bootstrap percentile and Bias-corrected accelerated (BCa) methods and simulation of the Likelihood Ratio Test lead to Confidence Intervals for Excess Relative Risk (ERR) and tests against the linear model. Results The linear model shows significant large, positive values of ERR for liver and urinary cancers at latencies from 37 - 43 years. Dose response below 20 mSv is strongly non-linear at the optimal latencies for the stomach (11.89 years), liver (36.9), lung (13.6), leukaemia (23.66), and pancreas (11.86) and across broad latency ranges. Confidence Intervals for ERR are comparable using Bootstrap and Likelihood Ratio Test methods and BCa 95% Confidence Intervals are strictly positive across latency ranges for all 5 cancers. Similar risk estimates for 10 mSv (lagged dose) are obtained from the 0 - 20 mSv and 5 - 500 mSv data for the stomach, liver, lung and leukaemia. Dose response for the latter 3 cancers is significantly non-linear in the 5 - 500 mSv range. Conclusion Liver and urinary cancer mortality risk is significantly raised using a latency model with linear dose response. A non-linear model is strongly superior for the stomach, liver, lung, pancreas and leukaemia. Bootstrap and Likelihood-based confidence intervals are broadly comparable and ERR is strictly positive by bootstrap methods for all 5 cancers. Except for the pancreas, similar estimates of latency and risk from 10 mSv are obtained from the 0 - 20 mSv and 5 - 500 mSv subcohorts. Large and significant cancer risks for Japanese survivors exposed to less than 20 mSv external radiation from the atomic bombs in 1945 cast doubt on the ICRP recommended annual occupational dose limit. PMID:20003238

A review and comparison of Bayesian and likelihood-based inferences in beta regression and zero-or-one-inflated beta regression.

PubMed

Liu, Fang; Eugenio, Evercita C

2018-04-01

Beta regression is an increasingly popular statistical technique in medical research for modeling of outcomes that assume values in (0, 1), such as proportions and patient reported outcomes. When outcomes take values in the intervals [0,1), (0,1], or [0,1], zero-or-one-inflated beta (zoib) regression can be used. We provide a thorough review on beta regression and zoib regression in the modeling, inferential, and computational aspects via the likelihood-based and Bayesian approaches. We demonstrate the statistical and practical importance of correctly modeling the inflation at zero/one rather than ad hoc replacing them with values close to zero/one via simulation studies; the latter approach can lead to biased estimates and invalid inferences. We show via simulation studies that the likelihood-based approach is computationally faster in general than MCMC algorithms used in the Bayesian inferences, but runs the risk of non-convergence, large biases, and sensitivity to starting values in the optimization algorithm especially with clustered/correlated data, data with sparse inflation at zero and one, and data that warrant regularization of the likelihood. The disadvantages of the regular likelihood-based approach make the Bayesian approach an attractive alternative in these cases. Software packages and tools for fitting beta and zoib regressions in both the likelihood-based and Bayesian frameworks are also reviewed.

Modeling short duration extreme precipitation patterns using copula and generalized maximum pseudo-likelihood estimation with censoring

NASA Astrophysics Data System (ADS)

Bargaoui, Zoubeida Kebaili; Bardossy, Andràs

2015-10-01

The paper aims to develop researches on the spatial variability of heavy rainfall events estimation using spatial copula analysis. To demonstrate the methodology, short time resolution rainfall time series from Stuttgart region are analyzed. They are constituted by rainfall observations on continuous 30 min time scale recorded over a network composed by 17 raingages for the period July 1989-July 2004. The analysis is performed aggregating the observations from 30 min up to 24 h. Two parametric bivariate extreme copula models, the Husler-Reiss model and the Gumbel model are investigated. Both involve a single parameter to be estimated. Thus, model fitting is operated for every pair of stations for a giving time resolution. A rainfall threshold value representing a fixed rainfall quantile is adopted for model inference. Generalized maximum pseudo-likelihood estimation is adopted with censoring by analogy with methods of univariate estimation combining historical and paleoflood information with systematic data. Only pairs of observations greater than the threshold are assumed as systematic data. Using the estimated copula parameter, a synthetic copula field is randomly generated and helps evaluating model adequacy which is achieved using Kolmogorov Smirnov distance test. In order to assess dependence or independence in the upper tail, the extremal coefficient which characterises the tail of the joint bivariate distribution is adopted. Hence, the extremal coefficient is reported as a function of the interdistance between stations. If it is less than 1.7, stations are interpreted as dependent in the extremes. The analysis of the fitted extremal coefficients with respect to stations inter distance highlights two regimes with different dependence structures: a short spatial extent regime linked to short duration intervals (from 30 min to 6 h) with an extent of about 8 km and a large spatial extent regime related to longer rainfall intervals (from 12 h to 24 h) with an extent of 34 to 38 km.

Estimating the variance for heterogeneity in arm-based network meta-analysis.

PubMed

Piepho, Hans-Peter; Madden, Laurence V; Roger, James; Payne, Roger; Williams, Emlyn R

2018-04-19

Network meta-analysis can be implemented by using arm-based or contrast-based models. Here we focus on arm-based models and fit them using generalized linear mixed model procedures. Full maximum likelihood (ML) estimation leads to biased trial-by-treatment interaction variance estimates for heterogeneity. Thus, our objective is to investigate alternative approaches to variance estimation that reduce bias compared with full ML. Specifically, we use penalized quasi-likelihood/pseudo-likelihood and hierarchical (h) likelihood approaches. In addition, we consider a novel model modification that yields estimators akin to the residual maximum likelihood estimator for linear mixed models. The proposed methods are compared by simulation, and 2 real datasets are used for illustration. Simulations show that penalized quasi-likelihood/pseudo-likelihood and h-likelihood reduce bias and yield satisfactory coverage rates. Sum-to-zero restriction and baseline contrasts for random trial-by-treatment interaction effects, as well as a residual ML-like adjustment, also reduce bias compared with an unconstrained model when ML is used, but coverage rates are not quite as good. Penalized quasi-likelihood/pseudo-likelihood and h-likelihood are therefore recommended. Copyright © 2018 John Wiley & Sons, Ltd.

Effects of mass media campaign exposure intensity and durability on quit attempts in a population-based cohort study.

PubMed

Wakefield, M A; Spittal, M J; Yong, H-H; Durkin, S J; Borland, R

2011-12-01

To assess the extent to which intensity and timing of televised anti-smoking advertising emphasizing the serious harms of smoking influences quit attempts. Using advertising gross rating points (GRPs), we estimated exposure to tobacco control and nicotine replacement therapy (NRT) advertising in the 3, 4-6, 7-9 and 10-12 months prior to follow-up of a replenished cohort of 3037 Australian smokers during 2002-08. Using generalized estimating equations, we related the intensity and timing of advertising exposure from each source to the likelihood of making a quit attempt in the 3 months prior to follow-up. Tobacco control advertising in the 3-month period prior to follow-up, but not in more distant past periods, was related to a higher likelihood of making a quit attempt. Each 1000 GRP increase per quarter was associated with an 11% increase in making a quit attempt [odds ratio (OR) = 1.11, 95% confidence interval (CI) 1.03-1.19, P = 0.009)]. NRT advertising was unrelated to quit attempts. Tobacco control advertising emphasizing the serious harms of smoking is associated with short-term increases in the likelihood of smokers making a quit attempt. Repeated cycles of higher intensity tobacco control media campaigns are needed to sustain high levels of quit attempts.

A hidden Markov model approach to neuron firing patterns.

PubMed

Camproux, A C; Saunier, F; Chouvet, G; Thalabard, J C; Thomas, G

1996-11-01

Analysis and characterization of neuronal discharge patterns are of interest to neurophysiologists and neuropharmacologists. In this paper we present a hidden Markov model approach to modeling single neuron electrical activity. Basically the model assumes that each interspike interval corresponds to one of several possible states of the neuron. Fitting the model to experimental series of interspike intervals by maximum likelihood allows estimation of the number of possible underlying neuron states, the probability density functions of interspike intervals corresponding to each state, and the transition probabilities between states. We present an application to the analysis of recordings of a locus coeruleus neuron under three pharmacological conditions. The model distinguishes two states during halothane anesthesia and during recovery from halothane anesthesia, and four states after administration of clonidine. The transition probabilities yield additional insights into the mechanisms of neuron firing.

Nonparametric change point estimation for survival distributions with a partially constant hazard rate.

PubMed

Brazzale, Alessandra R; Küchenhoff, Helmut; Krügel, Stefanie; Schiergens, Tobias S; Trentzsch, Heiko; Hartl, Wolfgang

2018-04-05

We present a new method for estimating a change point in the hazard function of a survival distribution assuming a constant hazard rate after the change point and a decreasing hazard rate before the change point. Our method is based on fitting a stump regression to p values for testing hazard rates in small time intervals. We present three real data examples describing survival patterns of severely ill patients, whose excess mortality rates are known to persist far beyond hospital discharge. For designing survival studies in these patients and for the definition of hospital performance metrics (e.g. mortality), it is essential to define adequate and objective end points. The reliable estimation of a change point will help researchers to identify such end points. By precisely knowing this change point, clinicians can distinguish between the acute phase with high hazard (time elapsed after admission and before the change point was reached), and the chronic phase (time elapsed after the change point) in which hazard is fairly constant. We show in an extensive simulation study that maximum likelihood estimation is not robust in this setting, and we evaluate our new estimation strategy including bootstrap confidence intervals and finite sample bias correction.

Refusal bias in HIV prevalence estimates from nationally representative seroprevalence surveys.

PubMed

Reniers, Georges; Eaton, Jeffrey

2009-03-13

To assess the relationship between prior knowledge of one's HIV status and the likelihood to refuse HIV testing in populations-based surveys and explore its potential for producing bias in HIV prevalence estimates. Using longitudinal survey data from Malawi, we estimate the relationship between prior knowledge of HIV-positive status and subsequent refusal of an HIV test. We use that parameter to develop a heuristic model of refusal bias that is applied to six Demographic and Health Surveys, in which refusal by HIV status is not observed. The model only adjusts for refusal bias conditional on a completed interview. Ecologically, HIV prevalence, prior testing rates and refusal for HIV testing are highly correlated. Malawian data further suggest that amongst individuals who know their status, HIV-positive individuals are 4.62 (95% confidence interval, 2.60-8.21) times more likely to refuse testing than HIV-negative ones. On the basis of that parameter and other inputs from the Demographic and Health Surveys, our model predicts downward bias in national HIV prevalence estimates ranging from 1.5% (95% confidence interval, 0.7-2.9) for Senegal to 13.3% (95% confidence interval, 7.2-19.6) for Malawi. In absolute terms, bias in HIV prevalence estimates is negligible for Senegal but 1.6 (95% confidence interval, 0.8-2.3) percentage points for Malawi. Downward bias is more severe in urban populations. Because refusal rates are higher in men, seroprevalence surveys also tend to overestimate the female-to-male ratio of infections. Prior knowledge of HIV status informs decisions to participate in seroprevalence surveys. Informed refusals may produce bias in estimates of HIV prevalence and the sex ratio of infections.

Highly variable recurrence of tsunamis in the 7,400 years before the 2004 Indian Ocean tsunami

NASA Astrophysics Data System (ADS)

Horton, B.; Rubin, C. M.; Sieh, K.; Jessica, P.; Daly, P.; Ismail, N.; Parnell, A. C.

2017-12-01

The devastating 2004 Indian Ocean tsunami caught millions of coastal residents and the scientific community off-guard. Subsequent research in the Indian Ocean basin has identified prehistoric tsunamis, but the timing and recurrence intervals of such events are uncertain. Here, we identify coastal caves as a new depositional environment for reconstructing tsunami records and present a 5,000 year record of continuous tsunami deposits from a coastal cave in Sumatra, Indonesia which shows the irregular recurrence of 11 tsunamis between 7,400 and 2,900 years BP. The data demonstrates that the 2004 tsunami was just the latest in a sequence of devastating tsunamis stretching back to at least the early Holocene and suggests a high likelihood for future tsunamis in the Indian Ocean. The sedimentary record in the cave shows that ruptures of the Sunda megathrust vary between large (which generated the 2004 Indian Ocean tsunami) and smaller slip failures. The chronology of events suggests the recurrence of multiple smaller tsunamis within relatively short time periods, interrupted by long periods of strain accumulation followed by giant tsunamis. The average time period between tsunamis is about 450 years with intervals ranging from a long, dormant period of over 2,000 years, to multiple tsunamis within the span of a century. The very long dormant period suggests that the Sunda megathrust is capable of accumulating large slip deficits between earthquakes. Such a high slip rupture would produce a substantially larger earthquake than the 2004 event. Although there is evidence that the likelihood of another tsunamigenic earthquake in Aceh province is high, these variable recurrence intervals suggest that long dormant periods may follow Sunda Megathrust ruptures as large as that of 2004 Indian Ocean tsunami. The remarkable variability of recurrence suggests that regional hazard mitigation plans should be based upon the high likelihood of future destructive tsunami demonstrated by the cave record and other paleotsunami sites, rather than estimates of recurrence intervals.

A study on reproductive performance and related factors of zebu cows in pastoral herds in a semi-arid area of Tanzania.

PubMed

Kanuya, N L; Matiko, M K; Kessy, B M; Mgongo, F O; Ropstad, E; Reksen, O

2006-06-01

A prospective longitudinal study was carried out from September 2001 to June 2004 in three adjacent villages in a semi-arid area of Tanzania. The objectives of this study were to measure the intervals between calving and either resumption of cyclical activity or confirmation of pregnancy, to estimate calving intervals, and to investigate the effect of factors assumed to be related to postpartum reproductive performance. A total of 275 lactation periods from 177 Tanzanian Shorthorn Zebu cows managed in a traditional pastoral system in 46 households were initially included. Animals were initially screened for brucelosis and thereafter examined by palpation per rectum at 2-week intervals. Body condition score (scale 1 to 5) was assessed and girth measurement (cm) taken. Occurrence of other reproductive events such as calving, abortion, death of calf, culling and reason for culling were recorded. In a subset of 98 lactation periods from 91 cows milk samples for progesterone (P4) determination were collected twice per week from day 7 after calving to the time of confirmed pregnancy or until milk production ceased before pregnancy. The data were analysed both univariately and in multivariable Cox proportional hazard (frailty) models. The mean (+/-S.E.M.) calving interval was 500+/-13.6 days. Positive reactors in the brucellosis test were 15.6% of the tested animals. Milk P4 analysis showed the rate of abortion/late embryo loss to be 14.3%. Calf mortality rates varied between 14.6 and 17.4%. A positive relationship was found between the outcome variables likelihood of cyclical activity and likelihood of pregnancy in the Cox model, and the explanatory variables: parity and body condition score (BCS) at calving. A negative relationship was found between the outcome variables, and the explanatory variables: maximum BCS loss and calf survival/mortality. Calving in the rainy season was associated with an increased likelihood of pregnancy.

Estimating numbers of greater prairie-chickens using mark-resight techniques

USGS Publications Warehouse

Clifton, A.M.; Krementz, D.G.

2006-01-01

Current monitoring efforts for greater prairie-chicken (Tympanuchus cupido pinnatus) populations indicate that populations are declining across their range. Monitoring the population status of greater prairie-chickens is based on traditional lek surveys (TLS) that provide an index without considering detectability. Estimators, such as immigration-emigration joint maximum-likelihood estimator from a hypergeometric distribution (IEJHE), can account for detectability and provide reliable population estimates based on resightings. We evaluated the use of mark-resight methods using radiotelemetry to estimate population size and density of greater prairie-chickens on 2 sites at a tallgrass prairie in the Flint Hills of Kansas, USA. We used average distances traveled from lek of capture to estimate density. Population estimates and confidence intervals at the 2 sites were 54 (CI 50-59) on 52.9 km 2 and 87 (CI 82-94) on 73.6 km2. The TLS performed at the same sites resulted in population ranges of 7-34 and 36-63 and always produced a lower population index than the mark-resight population estimate with a larger range. Mark-resight simulations with varying male:female ratios of marks indicated that this ratio was important in designing a population study on prairie-chickens. Confidence intervals for estimates when no marks were placed on females at the 2 sites (CI 46-50, 76-84) did not overlap confidence intervals when 40% of marks were placed on females (CI 54-64, 91-109). Population estimates derived using this mark-resight technique were apparently more accurate than traditional methods and would be more effective in detecting changes in prairie-chicken populations. Our technique could improve prairie-chicken management by providing wildlife biologists and land managers with a tool to estimate the population size and trends of lekking bird species, such as greater prairie-chickens.

Estimation of Model's Marginal likelihood Using Adaptive Sparse Grid Surrogates in Bayesian Model Averaging

NASA Astrophysics Data System (ADS)

Zeng, X.

2015-12-01

A large number of model executions are required to obtain alternative conceptual models' predictions and their posterior probabilities in Bayesian model averaging (BMA). The posterior model probability is estimated through models' marginal likelihood and prior probability. The heavy computation burden hinders the implementation of BMA prediction, especially for the elaborated marginal likelihood estimator. For overcoming the computation burden of BMA, an adaptive sparse grid (SG) stochastic collocation method is used to build surrogates for alternative conceptual models through the numerical experiment of a synthetical groundwater model. BMA predictions depend on model posterior weights (or marginal likelihoods), and this study also evaluated four marginal likelihood estimators, including arithmetic mean estimator (AME), harmonic mean estimator (HME), stabilized harmonic mean estimator (SHME), and thermodynamic integration estimator (TIE). The results demonstrate that TIE is accurate in estimating conceptual models' marginal likelihoods. The BMA-TIE has better predictive performance than other BMA predictions. TIE has high stability for estimating conceptual model's marginal likelihood. The repeated estimated conceptual model's marginal likelihoods by TIE have significant less variability than that estimated by other estimators. In addition, the SG surrogates are efficient to facilitate BMA predictions, especially for BMA-TIE. The number of model executions needed for building surrogates is 4.13%, 6.89%, 3.44%, and 0.43% of the required model executions of BMA-AME, BMA-HME, BMA-SHME, and BMA-TIE, respectively.

Quantifying Risk Over the Life Course – Latency, Age-Related Susceptibility, and Other Time-Varying Exposure Metrics

PubMed Central

Wang, Molin; Liao, Xiaomei; Laden, Francine; Spiegelman, Donna

2016-01-01

Identification of the latency period and age-related susceptibility, if any, is an important aspect of assessing risks of environmental, nutritional and occupational exposures. We consider estimation and inference for latency and age-related susceptibility in relative risk and excess risk models. We focus on likelihood-based methods for point and interval estimation of the latency period and age-related windows of susceptibility coupled with several commonly considered exposure metrics. The method is illustrated in a study of the timing of the effects of constituents of air pollution on mortality in the Nurses’ Health Study. PMID:26750582

Determinants of preterm birth rates in Canada from 1981 through 1983 and from 1992 through 1994.

PubMed

Joseph, K S; Kramer, M S; Marcoux, S; Ohlsson, A; Wen, S W; Allen, A; Platt, R

1998-11-12

The rates of preterm birth have increased in many countries, including Canada, over the past 20 years. However, the factors underlying the increase are poorly understood. We used data from the Statistics Canada live-birth and stillbirth data bases to determine the effects of changes in the frequency of multiple births, registration of births occurring very early in gestation, patterns of obstetrical intervention, and use of ultrasonographic dating of gestational age on the rates of preterm birth in Canada from 1981 through 1983 and from 1992 through 1994. All births in 9 of the 12 provinces and territories of Canada were included. Logistic-regression analysis and Poisson regression analysis were used to estimate changes between the two three-year periods, after adjustment for the above-mentioned determinants of the likelihood of preterm births. Preterm births increased from 6.3 percent of live births in 1981 through 1983 to 6.8 percent in 1992 through 1994, a relative increase of 9 percent (95 percent confidence interval, 7 to 10 percent). Among singleton births, preterm births increased by 5 percent (95 percent confidence interval, 3 to 6 percent). Multiple births increased from 1.9 percent to 2.1 percent of all live births; the rates of preterm birth among live births resulting from multiple gestations increased by 25 percent (95 percent confidence interval, 21 to 28 percent). Adjustment for the determinants of the likelihood of preterm birth reduced the increase in the rate of preterm birth to 3 percent among all live births and 1 percent among singleton births. The recent increase in preterm births in Canada is largely attributable to changes in the frequency of multiple births, obstetrical intervention, and the use of ultrasound-based estimates of gestational age.

Determining the accuracy of maximum likelihood parameter estimates with colored residuals

NASA Technical Reports Server (NTRS)

Morelli, Eugene A.; Klein, Vladislav

1994-01-01

An important part of building high fidelity mathematical models based on measured data is calculating the accuracy associated with statistical estimates of the model parameters. Indeed, without some idea of the accuracy of parameter estimates, the estimates themselves have limited value. In this work, an expression based on theoretical analysis was developed to properly compute parameter accuracy measures for maximum likelihood estimates with colored residuals. This result is important because experience from the analysis of measured data reveals that the residuals from maximum likelihood estimation are almost always colored. The calculations involved can be appended to conventional maximum likelihood estimation algorithms. Simulated data runs were used to show that the parameter accuracy measures computed with this technique accurately reflect the quality of the parameter estimates from maximum likelihood estimation without the need for analysis of the output residuals in the frequency domain or heuristically determined multiplication factors. The result is general, although the application studied here is maximum likelihood estimation of aerodynamic model parameters from flight test data.

On the existence of maximum likelihood estimates for presence-only data

USGS Publications Warehouse

Hefley, Trevor J.; Hooten, Mevin B.

2015-01-01

It is important to identify conditions for which maximum likelihood estimates are unlikely to be identifiable from presence-only data. In data sets where the maximum likelihood estimates do not exist, penalized likelihood and Bayesian methods will produce coefficient estimates, but these are sensitive to the choice of estimation procedure and prior or penalty term. When sample size is small or it is thought that habitat preferences are strong, we propose a suite of estimation procedures researchers can consider using.

Methods to estimate the between‐study variance and its uncertainty in meta‐analysis†

PubMed Central

Jackson, Dan; Viechtbauer, Wolfgang; Bender, Ralf; Bowden, Jack; Knapp, Guido; Kuss, Oliver; Higgins, Julian PT; Langan, Dean; Salanti, Georgia

2015-01-01

Meta‐analyses are typically used to estimate the overall/mean of an outcome of interest. However, inference about between‐study variability, which is typically modelled using a between‐study variance parameter, is usually an additional aim. The DerSimonian and Laird method, currently widely used by default to estimate the between‐study variance, has been long challenged. Our aim is to identify known methods for estimation of the between‐study variance and its corresponding uncertainty, and to summarise the simulation and empirical evidence that compares them. We identified 16 estimators for the between‐study variance, seven methods to calculate confidence intervals, and several comparative studies. Simulation studies suggest that for both dichotomous and continuous data the estimator proposed by Paule and Mandel and for continuous data the restricted maximum likelihood estimator are better alternatives to estimate the between‐study variance. Based on the scenarios and results presented in the published studies, we recommend the Q‐profile method and the alternative approach based on a ‘generalised Cochran between‐study variance statistic’ to compute corresponding confidence intervals around the resulting estimates. Our recommendations are based on a qualitative evaluation of the existing literature and expert consensus. Evidence‐based recommendations require an extensive simulation study where all methods would be compared under the same scenarios. © 2015 The Authors. Research Synthesis Methods published by John Wiley & Sons Ltd. PMID:26332144

Estimating Function Approaches for Spatial Point Processes

NASA Astrophysics Data System (ADS)

Deng, Chong

Spatial point pattern data consist of locations of events that are often of interest in biological and ecological studies. Such data are commonly viewed as a realization from a stochastic process called spatial point process. To fit a parametric spatial point process model to such data, likelihood-based methods have been widely studied. However, while maximum likelihood estimation is often too computationally intensive for Cox and cluster processes, pairwise likelihood methods such as composite likelihood, Palm likelihood usually suffer from the loss of information due to the ignorance of correlation among pairs. For many types of correlated data other than spatial point processes, when likelihood-based approaches are not desirable, estimating functions have been widely used for model fitting. In this dissertation, we explore the estimating function approaches for fitting spatial point process models. These approaches, which are based on the asymptotic optimal estimating function theories, can be used to incorporate the correlation among data and yield more efficient estimators. We conducted a series of studies to demonstrate that these estmating function approaches are good alternatives to balance the trade-off between computation complexity and estimating efficiency. First, we propose a new estimating procedure that improves the efficiency of pairwise composite likelihood method in estimating clustering parameters. Our approach combines estimating functions derived from pairwise composite likeli-hood estimation and estimating functions that account for correlations among the pairwise contributions. Our method can be used to fit a variety of parametric spatial point process models and can yield more efficient estimators for the clustering parameters than pairwise composite likelihood estimation. We demonstrate its efficacy through a simulation study and an application to the longleaf pine data. Second, we further explore the quasi-likelihood approach on fitting second-order intensity function of spatial point processes. However, the original second-order quasi-likelihood is barely feasible due to the intense computation and high memory requirement needed to solve a large linear system. Motivated by the existence of geometric regular patterns in the stationary point processes, we find a lower dimension representation of the optimal weight function and propose a reduced second-order quasi-likelihood approach. Through a simulation study, we show that the proposed method not only demonstrates superior performance in fitting the clustering parameter but also merits in the relaxation of the constraint of the tuning parameter, H. Third, we studied the quasi-likelihood type estimating funciton that is optimal in a certain class of first-order estimating functions for estimating the regression parameter in spatial point process models. Then, by using a novel spectral representation, we construct an implementation that is computationally much more efficient and can be applied to more general setup than the original quasi-likelihood method.

Provider perceptions concerning use of chest x-ray studies in adult blunt trauma assessments.

PubMed

Calderon, Georgina; Perez, Daniel; Fortman, Jonathan; Kea, Bory; Rodriguez, Robert M

2012-10-01

Although they infrequently lead to management changing diagnoses, chest x-rays (CXRs) are the most commonly ordered imaging study in blunt trauma evaluation. To determine: 1) the reasons physicians order chest X-ray studies (CXRs) in blunt trauma assessments; 2) what injuries they expect CXRs to reveal; and 3) whether physicians can accurately predict low likelihood of injury on CXR. At a Level I Trauma Center, we asked resident and attending physicians treating adult blunt trauma patients: 1) the primary reason(s) for getting CXRs; 2) what, if any, significant intrathoracic injuries (SITI) they expected CXRs to reveal; and 3) the likelihood of these injuries. An expert panel defined SITI as two or more rib fractures, sternal fracture, pulmonary contusion, pneumothorax, hemothorax, or aortic injury on official CXR readings. There were 484 patient encounters analyzed--65% of participating physicians were residents and 35% were attendings; 16 (3.3%) patients had SITI. The most common reasons for ordering CXRs were: "enough concern for significant injury" (62.9%) and belief that CXR is a "standard part of trauma work-up" (24.8%). Residents were more likely than attendings to cite "standard trauma work-up" (mean difference = 13.5%, p = 0.003). When physicians estimated a < 10% likelihood of SITI on CXR, 2.1% (95% confidence interval [CI] 1.0-4.1%) of patients had SITI; when they predicted a 10-25% likelihood, 5.7% (95% CI 1.2-15.7%) had SITI; and when they predicted a > 25% likelihood, 9.1% (95% CI 3.0-20.0%) had SITI. Physicians order CXRs in blunt trauma patients because they expect to find injuries and believe that CXRs are part of a "standard" work-up. Providers commonly do not expect CXRs to reveal SITI. When providers estimated low likelihood of SITI, the rate of SITI was very low. Published by Elsevier Inc.

Modeling gene expression measurement error: a quasi-likelihood approach

PubMed Central

Strimmer, Korbinian

2003-01-01

Background Using suitable error models for gene expression measurements is essential in the statistical analysis of microarray data. However, the true probabilistic model underlying gene expression intensity readings is generally not known. Instead, in currently used approaches some simple parametric model is assumed (usually a transformed normal distribution) or the empirical distribution is estimated. However, both these strategies may not be optimal for gene expression data, as the non-parametric approach ignores known structural information whereas the fully parametric models run the risk of misspecification. A further related problem is the choice of a suitable scale for the model (e.g. observed vs. log-scale). Results Here a simple semi-parametric model for gene expression measurement error is presented. In this approach inference is based an approximate likelihood function (the extended quasi-likelihood). Only partial knowledge about the unknown true distribution is required to construct this function. In case of gene expression this information is available in the form of the postulated (e.g. quadratic) variance structure of the data. As the quasi-likelihood behaves (almost) like a proper likelihood, it allows for the estimation of calibration and variance parameters, and it is also straightforward to obtain corresponding approximate confidence intervals. Unlike most other frameworks, it also allows analysis on any preferred scale, i.e. both on the original linear scale as well as on a transformed scale. It can also be employed in regression approaches to model systematic (e.g. array or dye) effects. Conclusions The quasi-likelihood framework provides a simple and versatile approach to analyze gene expression data that does not make any strong distributional assumptions about the underlying error model. For several simulated as well as real data sets it provides a better fit to the data than competing models. In an example it also improved the power of tests to identify differential expression. PMID:12659637

Using Stochastic Approximation Techniques to Efficiently Construct Confidence Intervals for Heritability.

PubMed

Schweiger, Regev; Fisher, Eyal; Rahmani, Elior; Shenhav, Liat; Rosset, Saharon; Halperin, Eran

2018-06-22

Estimation of heritability is an important task in genetics. The use of linear mixed models (LMMs) to determine narrow-sense single-nucleotide polymorphism (SNP)-heritability and related quantities has received much recent attention, due of its ability to account for variants with small effect sizes. Typically, heritability estimation under LMMs uses the restricted maximum likelihood (REML) approach. The common way to report the uncertainty in REML estimation uses standard errors (SEs), which rely on asymptotic properties. However, these assumptions are often violated because of the bounded parameter space, statistical dependencies, and limited sample size, leading to biased estimates and inflated or deflated confidence intervals (CIs). In addition, for larger data sets (e.g., tens of thousands of individuals), the construction of SEs itself may require considerable time, as it requires expensive matrix inversions and multiplications. Here, we present FIESTA (Fast confidence IntErvals using STochastic Approximation), a method for constructing accurate CIs. FIESTA is based on parametric bootstrap sampling, and, therefore, avoids unjustified assumptions on the distribution of the heritability estimator. FIESTA uses stochastic approximation techniques, which accelerate the construction of CIs by several orders of magnitude, compared with previous approaches as well as to the analytical approximation used by SEs. FIESTA builds accurate CIs rapidly, for example, requiring only several seconds for data sets of tens of thousands of individuals, making FIESTA a very fast solution to the problem of building accurate CIs for heritability for all data set sizes.

Finite mixture model: A maximum likelihood estimation approach on time series data

NASA Astrophysics Data System (ADS)

Yen, Phoong Seuk; Ismail, Mohd Tahir; Hamzah, Firdaus Mohamad

2014-09-01

Recently, statistician emphasized on the fitting of finite mixture model by using maximum likelihood estimation as it provides asymptotic properties. In addition, it shows consistency properties as the sample sizes increases to infinity. This illustrated that maximum likelihood estimation is an unbiased estimator. Moreover, the estimate parameters obtained from the application of maximum likelihood estimation have smallest variance as compared to others statistical method as the sample sizes increases. Thus, maximum likelihood estimation is adopted in this paper to fit the two-component mixture model in order to explore the relationship between rubber price and exchange rate for Malaysia, Thailand, Philippines and Indonesia. Results described that there is a negative effect among rubber price and exchange rate for all selected countries.

A hidden Markov model approach to neuron firing patterns.

PubMed Central

Camproux, A C; Saunier, F; Chouvet, G; Thalabard, J C; Thomas, G

1996-01-01

Analysis and characterization of neuronal discharge patterns are of interest to neurophysiologists and neuropharmacologists. In this paper we present a hidden Markov model approach to modeling single neuron electrical activity. Basically the model assumes that each interspike interval corresponds to one of several possible states of the neuron. Fitting the model to experimental series of interspike intervals by maximum likelihood allows estimation of the number of possible underlying neuron states, the probability density functions of interspike intervals corresponding to each state, and the transition probabilities between states. We present an application to the analysis of recordings of a locus coeruleus neuron under three pharmacological conditions. The model distinguishes two states during halothane anesthesia and during recovery from halothane anesthesia, and four states after administration of clonidine. The transition probabilities yield additional insights into the mechanisms of neuron firing. Images FIGURE 3 PMID:8913581

Calibration of two complex ecosystem models with different likelihood functions

NASA Astrophysics Data System (ADS)

Hidy, Dóra; Haszpra, László; Pintér, Krisztina; Nagy, Zoltán; Barcza, Zoltán

2014-05-01

The biosphere is a sensitive carbon reservoir. Terrestrial ecosystems were approximately carbon neutral during the past centuries, but they became net carbon sinks due to climate change induced environmental change and associated CO2 fertilization effect of the atmosphere. Model studies and measurements indicate that the biospheric carbon sink can saturate in the future due to ongoing climate change which can act as a positive feedback. Robustness of carbon cycle models is a key issue when trying to choose the appropriate model for decision support. The input parameters of the process-based models are decisive regarding the model output. At the same time there are several input parameters for which accurate values are hard to obtain directly from experiments or no local measurements are available. Due to the uncertainty associated with the unknown model parameters significant bias can be experienced if the model is used to simulate the carbon and nitrogen cycle components of different ecosystems. In order to improve model performance the unknown model parameters has to be estimated. We developed a multi-objective, two-step calibration method based on Bayesian approach in order to estimate the unknown parameters of PaSim and Biome-BGC models. Biome-BGC and PaSim are a widely used biogeochemical models that simulate the storage and flux of water, carbon, and nitrogen between the ecosystem and the atmosphere, and within the components of the terrestrial ecosystems (in this research the developed version of Biome-BGC is used which is referred as BBGC MuSo). Both models were calibrated regardless the simulated processes and type of model parameters. The calibration procedure is based on the comparison of measured data with simulated results via calculating a likelihood function (degree of goodness-of-fit between simulated and measured data). In our research different likelihood function formulations were used in order to examine the effect of the different model goodness metric on calibration. The different likelihoods are different functions of RMSE (root mean squared error) weighted by measurement uncertainty: exponential / linear / quadratic / linear normalized by correlation. As a first calibration step sensitivity analysis was performed in order to select the influential parameters which have strong effect on the output data. In the second calibration step only the sensitive parameters were calibrated (optimal values and confidence intervals were calculated). In case of PaSim more parameters were found responsible for the 95% of the output data variance than is case of BBGC MuSo. Analysis of the results of the optimized models revealed that the exponential likelihood estimation proved to be the most robust (best model simulation with optimized parameter, highest confidence interval increase). The cross-validation of the model simulations can help in constraining the highly uncertain greenhouse gas budget of grasslands.

A methodology for airplane parameter estimation and confidence interval determination in nonlinear estimation problems. Ph.D. Thesis - George Washington Univ., Apr. 1985

NASA Technical Reports Server (NTRS)

Murphy, P. C.

1986-01-01

An algorithm for maximum likelihood (ML) estimation is developed with an efficient method for approximating the sensitivities. The ML algorithm relies on a new optimization method referred to as a modified Newton-Raphson with estimated sensitivities (MNRES). MNRES determines sensitivities by using slope information from local surface approximations of each output variable in parameter space. With the fitted surface, sensitivity information can be updated at each iteration with less computational effort than that required by either a finite-difference method or integration of the analytically determined sensitivity equations. MNRES eliminates the need to derive sensitivity equations for each new model, and thus provides flexibility to use model equations in any convenient format. A random search technique for determining the confidence limits of ML parameter estimates is applied to nonlinear estimation problems for airplanes. The confidence intervals obtained by the search are compared with Cramer-Rao (CR) bounds at the same confidence level. The degree of nonlinearity in the estimation problem is an important factor in the relationship between CR bounds and the error bounds determined by the search technique. Beale's measure of nonlinearity is developed in this study for airplane identification problems; it is used to empirically correct confidence levels and to predict the degree of agreement between CR bounds and search estimates.

An algorithm for computing moments-based flood quantile estimates when historical flood information is available

USGS Publications Warehouse

Cohn, T.A.; Lane, W.L.; Baier, W.G.

1997-01-01

This paper presents the expected moments algorithm (EMA), a simple and efficient method for incorporating historical and paleoflood information into flood frequency studies. EMA can utilize three types of at-site flood information: systematic stream gage record; information about the magnitude of historical floods; and knowledge of the number of years in the historical period when no large flood occurred. EMA employs an iterative procedure to compute method-of-moments parameter estimates. Initial parameter estimates are calculated from systematic stream gage data. These moments are then updated by including the measured historical peaks and the expected moments, given the previously estimated parameters, of the below-threshold floods from the historical period. The updated moments result in new parameter estimates, and the last two steps are repeated until the algorithm converges. Monte Carlo simulations compare EMA, Bulletin 17B's [United States Water Resources Council, 1982] historically weighted moments adjustment, and maximum likelihood estimators when fitting the three parameters of the log-Pearson type III distribution. These simulations demonstrate that EMA is more efficient than the Bulletin 17B method, and that it is nearly as efficient as maximum likelihood estimation (MLE). The experiments also suggest that EMA has two advantages over MLE when dealing with the log-Pearson type III distribution: It appears that EMA estimates always exist and that they are unique, although neither result has been proven. EMA can be used with binomial or interval-censored data and with any distributional family amenable to method-of-moments estimation.

An algorithm for computing moments-based flood quantile estimates when historical flood information is available

NASA Astrophysics Data System (ADS)

Cohn, T. A.; Lane, W. L.; Baier, W. G.

This paper presents the expected moments algorithm (EMA), a simple and efficient method for incorporating historical and paleoflood information into flood frequency studies. EMA can utilize three types of at-site flood information: systematic stream gage record; information about the magnitude of historical floods; and knowledge of the number of years in the historical period when no large flood occurred. EMA employs an iterative procedure to compute method-of-moments parameter estimates. Initial parameter estimates are calculated from systematic stream gage data. These moments are then updated by including the measured historical peaks and the expected moments, given the previously estimated parameters, of the below-threshold floods from the historical period. The updated moments result in new parameter estimates, and the last two steps are repeated until the algorithm converges. Monte Carlo simulations compare EMA, Bulletin 17B's [United States Water Resources Council, 1982] historically weighted moments adjustment, and maximum likelihood estimators when fitting the three parameters of the log-Pearson type III distribution. These simulations demonstrate that EMA is more efficient than the Bulletin 17B method, and that it is nearly as efficient as maximum likelihood estimation (MLE). The experiments also suggest that EMA has two advantages over MLE when dealing with the log-Pearson type III distribution: It appears that EMA estimates always exist and that they are unique, although neither result has been proven. EMA can be used with binomial or interval-censored data and with any distributional family amenable to method-of-moments estimation.

Statistical inference for extended or shortened phase II studies based on Simon's two-stage designs.

PubMed

Zhao, Junjun; Yu, Menggang; Feng, Xi-Ping

2015-06-07

Simon's two-stage designs are popular choices for conducting phase II clinical trials, especially in the oncology trials to reduce the number of patients placed on ineffective experimental therapies. Recently Koyama and Chen (2008) discussed how to conduct proper inference for such studies because they found that inference procedures used with Simon's designs almost always ignore the actual sampling plan used. In particular, they proposed an inference method for studies when the actual second stage sample sizes differ from planned ones. We consider an alternative inference method based on likelihood ratio. In particular, we order permissible sample paths under Simon's two-stage designs using their corresponding conditional likelihood. In this way, we can calculate p-values using the common definition: the probability of obtaining a test statistic value at least as extreme as that observed under the null hypothesis. In addition to providing inference for a couple of scenarios where Koyama and Chen's method can be difficult to apply, the resulting estimate based on our method appears to have certain advantage in terms of inference properties in many numerical simulations. It generally led to smaller biases and narrower confidence intervals while maintaining similar coverages. We also illustrated the two methods in a real data setting. Inference procedures used with Simon's designs almost always ignore the actual sampling plan. Reported P-values, point estimates and confidence intervals for the response rate are not usually adjusted for the design's adaptiveness. Proper statistical inference procedures should be used.

Calculation of Weibull strength parameters and Batdorf flow-density constants for volume- and surface-flaw-induced fracture in ceramics

NASA Technical Reports Server (NTRS)

Pai, Shantaram S.; Gyekenyesi, John P.

1988-01-01

The calculation of shape and scale parameters of the two-parameter Weibull distribution is described using the least-squares analysis and maximum likelihood methods for volume- and surface-flaw-induced fracture in ceramics with complete and censored samples. Detailed procedures are given for evaluating 90 percent confidence intervals for maximum likelihood estimates of shape and scale parameters, the unbiased estimates of the shape parameters, and the Weibull mean values and corresponding standard deviations. Furthermore, the necessary steps are described for detecting outliers and for calculating the Kolmogorov-Smirnov and the Anderson-Darling goodness-of-fit statistics and 90 percent confidence bands about the Weibull distribution. It also shows how to calculate the Batdorf flaw-density constants by uing the Weibull distribution statistical parameters. The techniques described were verified with several example problems, from the open literature, and were coded. The techniques described were verified with several example problems from the open literature, and were coded in the Structural Ceramics Analysis and Reliability Evaluation (SCARE) design program.

Comparing the Performance of Improved Classify-Analyze Approaches For Distal Outcomes in Latent Profile Analysis

PubMed Central

Dziak, John J.; Bray, Bethany C.; Zhang, Jieting; Zhang, Minqiang; Lanza, Stephanie T.

2016-01-01

Several approaches are available for estimating the relationship of latent class membership to distal outcomes in latent profile analysis (LPA). A three-step approach is commonly used, but has problems with estimation bias and confidence interval coverage. Proposed improvements include the correction method of Bolck, Croon, and Hagenaars (BCH; 2004), Vermunt’s (2010) maximum likelihood (ML) approach, and the inclusive three-step approach of Bray, Lanza, & Tan (2015). These methods have been studied in the related case of latent class analysis (LCA) with categorical indicators, but not as well studied for LPA with continuous indicators. We investigated the performance of these approaches in LPA with normally distributed indicators, under different conditions of distal outcome distribution, class measurement quality, relative latent class size, and strength of association between latent class and the distal outcome. The modified BCH implemented in Latent GOLD had excellent performance. The maximum likelihood and inclusive approaches were not robust to violations of distributional assumptions. These findings broadly agree with and extend the results presented by Bakk and Vermunt (2016) in the context of LCA with categorical indicators. PMID:28630602

Software for Quantifying and Simulating Microsatellite Genotyping Error

PubMed Central

Johnson, Paul C.D.; Haydon, Daniel T.

2007-01-01

Microsatellite genetic marker data are exploited in a variety of fields, including forensics, gene mapping, kinship inference and population genetics. In all of these fields, inference can be thwarted by failure to quantify and account for data errors, and kinship inference in particular can benefit from separating errors into two distinct classes: allelic dropout and false alleles. Pedant is MS Windows software for estimating locus-specific maximum likelihood rates of these two classes of error. Estimation is based on comparison of duplicate error-prone genotypes: neither reference genotypes nor pedigree data are required. Other functions include: plotting of error rate estimates and confidence intervals; simulations for performing power analysis and for testing the robustness of error rate estimates to violation of the underlying assumptions; and estimation of expected heterozygosity, which is a required input. The program, documentation and source code are available from http://www.stats.gla.ac.uk/~paulj/pedant.html. PMID:20066126

Defining thresholds of specific IgE levels to grass pollen and birch pollen allergens improves clinical interpretation.

PubMed

Van Hoeyveld, Erna; Nickmans, Silvie; Ceuppens, Jan L; Bossuyt, Xavier

2015-10-23

Cut-off values and predictive values are used for the clinical interpretation of specific IgE antibody results. However, cut-off levels are not well defined, and predictive values are dependent on the prevalence of disease. The objective of this study was to document clinically relevant diagnostic accuracy of specific IgE for inhalant allergens (grass pollen and birch pollen) based on test result interval-specific likelihood ratios. Likelihood ratios are independent of the prevalence and allow to provide diagnostic accuracy information for test result intervals. In a prospective study we included consecutive adult patients presenting at an allergy clinic with complaints of rhinitis or rhinoconjunctivitis. The standard for diagnosis was a suggestive clinical history of grass or birch pollen allergy and a positive skin test. Specific IgE was determined with the ImmunoCAP Fluorescence Enzyme Immuno-Assay. We established specific IgE test result interval related likelihood ratios for clinical allergy to inhalant allergens (grass pollen, rPhl p 1,5, birch pollen, rBet v 1). The likelihood ratios for allergy increased with increasing specific IgE antibody levels. The likelihood ratio was <0.03 for specific IgE <0.1 kU/L, between 0.1 and 1.4 for specific IgE between 0.1 kU/L and 0.35 kU/L, between 1.4 and 4.2 for specific IgE between 0.35 kU/L and 3.5 kU/L, >6.3 for specific IgE>0.7, and very high (∞) for specific IgE >3.5 kU/L. Test result interval specific likelihood ratios provide a useful tool for the interpretation of specific IgE test results for inhalant allergens. Copyright © 2015 Elsevier B.V. All rights reserved.

Quantifying risk over the life course - latency, age-related susceptibility, and other time-varying exposure metrics.

PubMed

Wang, Molin; Liao, Xiaomei; Laden, Francine; Spiegelman, Donna

2016-06-15

Identification of the latency period and age-related susceptibility, if any, is an important aspect of assessing risks of environmental, nutritional, and occupational exposures. We consider estimation and inference for latency and age-related susceptibility in relative risk and excess risk models. We focus on likelihood-based methods for point and interval estimation of the latency period and age-related windows of susceptibility coupled with several commonly considered exposure metrics. The method is illustrated in a study of the timing of the effects of constituents of air pollution on mortality in the Nurses' Health Study. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

Effects of mass media campaign exposure intensity and durability on quit attempts in a population-based cohort study

PubMed Central

Wakefield, M. A.; Spittal, M. J.; Yong, H-H.; Durkin, S. J.; Borland, R.

2011-01-01

Objective: To assess the extent to which intensity and timing of televised anti-smoking advertising emphasizing the serious harms of smoking influences quit attempts. Methods: Using advertising gross rating points (GRPs), we estimated exposure to tobacco control and nicotine replacement therapy (NRT) advertising in the 3, 4–6, 7–9 and 10–12 months prior to follow-up of a replenished cohort of 3037 Australian smokers during 2002–08. Using generalized estimating equations, we related the intensity and timing of advertising exposure from each source to the likelihood of making a quit attempt in the 3 months prior to follow-up. Results: Tobacco control advertising in the 3-month period prior to follow-up, but not in more distant past periods, was related to a higher likelihood of making a quit attempt. Each 1000 GRP increase per quarter was associated with an 11% increase in making a quit attempt [odds ratio (OR) = 1.11, 95% confidence interval (CI) 1.03–1.19, P = 0.009)]. NRT advertising was unrelated to quit attempts. Conclusions: Tobacco control advertising emphasizing the serious harms of smoking is associated with short-term increases in the likelihood of smokers making a quit attempt. Repeated cycles of higher intensity tobacco control media campaigns are needed to sustain high levels of quit attempts. PMID:21730252

The numerical evaluation of maximum-likelihood estimates of the parameters for a mixture of normal distributions from partially identified samples

NASA Technical Reports Server (NTRS)

Walker, H. F.

1976-01-01

Likelihood equations determined by the two types of samples which are necessary conditions for a maximum-likelihood estimate were considered. These equations suggest certain successive approximations iterative procedures for obtaining maximum likelihood estimates. The procedures, which are generalized steepest ascent (deflected gradient) procedures, contain those of Hosmer as a special case.

Exposure to traffic-related air pollution during pregnancy and term low birth weight: estimation of causal associations in a semiparametric model.

PubMed

Padula, Amy M; Mortimer, Kathleen; Hubbard, Alan; Lurmann, Frederick; Jerrett, Michael; Tager, Ira B

2012-11-01

Traffic-related air pollution is recognized as an important contributor to health problems. Epidemiologic analyses suggest that prenatal exposure to traffic-related air pollutants may be associated with adverse birth outcomes; however, there is insufficient evidence to conclude that the relation is causal. The Study of Air Pollution, Genetics and Early Life Events comprises all births to women living in 4 counties in California's San Joaquin Valley during the years 2000-2006. The probability of low birth weight among full-term infants in the population was estimated using machine learning and targeted maximum likelihood estimation for each quartile of traffic exposure during pregnancy. If everyone lived near high-volume freeways (approximated as the fourth quartile of traffic density), the estimated probability of term low birth weight would be 2.27% (95% confidence interval: 2.16, 2.38) as compared with 2.02% (95% confidence interval: 1.90, 2.12) if everyone lived near smaller local roads (first quartile of traffic density). Assessment of potentially causal associations, in the absence of arbitrary model assumptions applied to the data, should result in relatively unbiased estimates. The current results support findings from previous studies that prenatal exposure to traffic-related air pollution may adversely affect birth weight among full-term infants.

Correcting for bias in the selection and validation of informative diagnostic tests.

PubMed

Robertson, David S; Prevost, A Toby; Bowden, Jack

2015-04-15

When developing a new diagnostic test for a disease, there are often multiple candidate classifiers to choose from, and it is unclear if any will offer an improvement in performance compared with current technology. A two-stage design can be used to select a promising classifier (if one exists) in stage one for definitive validation in stage two. However, estimating the true properties of the chosen classifier is complicated by the first stage selection rules. In particular, the usual maximum likelihood estimator (MLE) that combines data from both stages will be biased high. Consequently, confidence intervals and p-values flowing from the MLE will also be incorrect. Building on the results of Pepe et al. (SIM 28:762-779), we derive the most efficient conditionally unbiased estimator and exact confidence intervals for a classifier's sensitivity in a two-stage design with arbitrary selection rules; the condition being that the trial proceeds to the validation stage. We apply our estimation strategy to data from a recent family history screening tool validation study by Walter et al. (BJGP 63:393-400) and are able to identify and successfully adjust for bias in the tool's estimated sensitivity to detect those at higher risk of breast cancer. © 2015 The Authors. Statistics in Medicine published by John Wiley & Sons Ltd.

Semiparametric regression analysis of interval-censored competing risks data.

PubMed

Mao, Lu; Lin, Dan-Yu; Zeng, Donglin

2017-09-01

Interval-censored competing risks data arise when each study subject may experience an event or failure from one of several causes and the failure time is not observed directly but rather is known to lie in an interval between two examinations. We formulate the effects of possibly time-varying (external) covariates on the cumulative incidence or sub-distribution function of competing risks (i.e., the marginal probability of failure from a specific cause) through a broad class of semiparametric regression models that captures both proportional and non-proportional hazards structures for the sub-distribution. We allow each subject to have an arbitrary number of examinations and accommodate missing information on the cause of failure. We consider nonparametric maximum likelihood estimation and devise a fast and stable EM-type algorithm for its computation. We then establish the consistency, asymptotic normality, and semiparametric efficiency of the resulting estimators for the regression parameters by appealing to modern empirical process theory. In addition, we show through extensive simulation studies that the proposed methods perform well in realistic situations. Finally, we provide an application to a study on HIV-1 infection with different viral subtypes. © 2017, The International Biometric Society.

The performance of blood pressure-to-height ratio as a screening measure for identifying children and adolescents with hypertension: a meta-analysis.

PubMed

Ma, Chunming; Liu, Yue; Lu, Qiang; Lu, Na; Liu, Xiaoli; Tian, Yiming; Wang, Rui; Yin, Fuzai

2016-02-01

The blood pressure-to-height ratio (BPHR) has been shown to be an accurate index for screening hypertension in children and adolescents. The aim of the present study was to perform a meta-analysis to assess the performance of BPHR for the assessment of hypertension. Electronic and manual searches were performed to identify studies of the BPHR. After methodological quality assessment and data extraction, pooled estimates of the sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, area under the receiver operating characteristic curve and summary receiver operating characteristics were assessed systematically. The extent of heterogeneity for it was assessed. Six studies were identified for analysis. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and diagnostic odds ratio values of BPHR, for assessment of hypertension, were 96% [95% confidence interval (CI)=0.95-0.97], 90% (95% CI=0.90-0.91), 10.68 (95% CI=8.03-14.21), 0.04 (95% CI=0.03-0.07) and 247.82 (95% CI=114.50-536.34), respectively. The area under the receiver operating characteristic curve was 0.9472. The BPHR had higher diagnostic accuracies for identifying hypertension in children and adolescents.

Mean cerebral blood volume is an effective diagnostic index of recurrent and radiation injury in glioma patients: A meta-analysis of diagnostic test.

PubMed

Li, Zhanzhan; Zhou, Qin; Li, Yanyan; Yan, Shipeng; Fu, Jun; Huang, Xinqiong; Shen, Liangfang

2017-02-28

We conducted a meta-analysis to evaluate the diagnostic values of mean cerebral blood volume for recurrent and radiation injury in glioma patients. We performed systematic electronic searches for eligible study up to August 8, 2016. Bivariate mixed effects models were used to estimate the combined sensitivity, specificity, positive likelihood ratios, negative likelihood ratios, diagnostic odds ratios and their 95% confidence intervals (CIs). Fifteen studies with a total number of 576 participants were enrolled. The pooled sensitivity and specificity of diagnostic were 0.88 (95%CI: 0.82-0.92) and 0.85 (95%CI: 0.68-0.93). The pooled positive likelihood ratio is 5.73 (95%CI: 2.56-12.81), negative likelihood ratio is 0.15 (95%CI: 0.10-0.22), and the diagnostic odds ratio is 39.34 (95%CI:13.96-110.84). The summary receiver operator characteristic is 0.91 (95%CI: 0.88-0.93). However, the Deek's plot suggested publication bias may exist (t=2.30, P=0.039). Mean cerebral blood volume measurement methods seems to be very sensitive and highly specific to differentiate recurrent and radiation injury in glioma patients. The results should be interpreted with caution because of the potential bias.

Bias correction in the hierarchical likelihood approach to the analysis of multivariate survival data.

PubMed

Jeon, Jihyoun; Hsu, Li; Gorfine, Malka

2012-07-01

Frailty models are useful for measuring unobserved heterogeneity in risk of failures across clusters, providing cluster-specific risk prediction. In a frailty model, the latent frailties shared by members within a cluster are assumed to act multiplicatively on the hazard function. In order to obtain parameter and frailty variate estimates, we consider the hierarchical likelihood (H-likelihood) approach (Ha, Lee and Song, 2001. Hierarchical-likelihood approach for frailty models. Biometrika 88, 233-243) in which the latent frailties are treated as "parameters" and estimated jointly with other parameters of interest. We find that the H-likelihood estimators perform well when the censoring rate is low, however, they are substantially biased when the censoring rate is moderate to high. In this paper, we propose a simple and easy-to-implement bias correction method for the H-likelihood estimators under a shared frailty model. We also extend the method to a multivariate frailty model, which incorporates complex dependence structure within clusters. We conduct an extensive simulation study and show that the proposed approach performs very well for censoring rates as high as 80%. We also illustrate the method with a breast cancer data set. Since the H-likelihood is the same as the penalized likelihood function, the proposed bias correction method is also applicable to the penalized likelihood estimators.

Monte Carlo Method for Determining Earthquake Recurrence Parameters from Short Paleoseismic Catalogs: Example Calculations for California

USGS Publications Warehouse

Parsons, Tom

2008-01-01

Paleoearthquake observations often lack enough events at a given site to directly define a probability density function (PDF) for earthquake recurrence. Sites with fewer than 10-15 intervals do not provide enough information to reliably determine the shape of the PDF using standard maximum-likelihood techniques [e.g., Ellsworth et al., 1999]. In this paper I present a method that attempts to fit wide ranges of distribution parameters to short paleoseismic series. From repeated Monte Carlo draws, it becomes possible to quantitatively estimate most likely recurrence PDF parameters, and a ranked distribution of parameters is returned that can be used to assess uncertainties in hazard calculations. In tests on short synthetic earthquake series, the method gives results that cluster around the mean of the input distribution, whereas maximum likelihood methods return the sample means [e.g., NIST/SEMATECH, 2006]. For short series (fewer than 10 intervals), sample means tend to reflect the median of an asymmetric recurrence distribution, possibly leading to an overestimate of the hazard should they be used in probability calculations. Therefore a Monte Carlo approach may be useful for assessing recurrence from limited paleoearthquake records. Further, the degree of functional dependence among parameters like mean recurrence interval and coefficient of variation can be established. The method is described for use with time-independent and time-dependent PDF?s, and results from 19 paleoseismic sequences on strike-slip faults throughout the state of California are given.

Monte Carlo method for determining earthquake recurrence parameters from short paleoseismic catalogs: Example calculations for California

USGS Publications Warehouse

Parsons, T.

2008-01-01

Paleoearthquake observations often lack enough events at a given site to directly define a probability density function (PDF) for earthquake recurrence. Sites with fewer than 10-15 intervals do not provide enough information to reliably determine the shape of the PDF using standard maximum-likelihood techniques (e.g., Ellsworth et al., 1999). In this paper I present a method that attempts to fit wide ranges of distribution parameters to short paleoseismic series. From repeated Monte Carlo draws, it becomes possible to quantitatively estimate most likely recurrence PDF parameters, and a ranked distribution of parameters is returned that can be used to assess uncertainties in hazard calculations. In tests on short synthetic earthquake series, the method gives results that cluster around the mean of the input distribution, whereas maximum likelihood methods return the sample means (e.g., NIST/SEMATECH, 2006). For short series (fewer than 10 intervals), sample means tend to reflect the median of an asymmetric recurrence distribution, possibly leading to an overestimate of the hazard should they be used in probability calculations. Therefore a Monte Carlo approach may be useful for assessing recurrence from limited paleoearthquake records. Further, the degree of functional dependence among parameters like mean recurrence interval and coefficient of variation can be established. The method is described for use with time-independent and time-dependent PDFs, and results from 19 paleoseismic sequences on strike-slip faults throughout the state of California are given.

The logistic model for predicting the non-gonoactive Aedes aegypti females.

PubMed

Reyes-Villanueva, Filiberto; Rodríguez-Pérez, Mario A

2004-01-01

To estimate, using logistic regression, the likelihood of occurrence of a non-gonoactive Aedes aegypti female, previously fed human blood, with relation to body size and collection method. This study was conducted in Monterrey, Mexico, between 1994 and 1996. Ten samplings of 60 mosquitoes of Ae. aegypti females were carried out in three dengue endemic areas: six of biting females, two of emerging mosquitoes, and two of indoor resting females. Gravid females, as well as those with blood in the gut were removed. Mosquitoes were taken to the laboratory and engorged on human blood. After 48 hours, ovaries were dissected to register whether they were gonoactive or non-gonoactive. Wing-length in mm was an indicator for body size. The logistic regression model was used to assess the likelihood of non-gonoactivity, as a binary variable, in relation to wing-length and collection method. Of the 600 females, 164 (27%) remained non-gonoactive, with a wing-length range of 1.9-3.2 mm, almost equal to that of all females (1.8-3.3 mm). The logistic regression model showed a significant likelihood of a female remaining non-gonoactive (Y=1). The collection method did not influence the binary response, but there was an inverse relationship between non-gonoactivity and wing-length. Dengue vector populations from Monterrey, Mexico display a wide-range body size. Logistic regression was a useful tool to estimate the likelihood for an engorged female to remain non-gonoactive. The necessity for a second blood meal is present in any female, but small mosquitoes are more likely to bite again within a 2-day interval, in order to attain egg maturation. The English version of this paper is available too at: http://www.insp.mx/salud/index.html.

Introducing Computed Tomography Standards for Age Estimation of Modern Australian Subadults Using Postnatal Ossification Timings of Select Cranial and Cervical Sites(.).

PubMed

Lottering, Nicolene; MacGregor, Donna M; Alston, Clair L; Watson, Debbie; Gregory, Laura S

2016-01-01

Contemporary, population-specific ossification timings of the cranium are lacking in current literature due to challenges in obtaining large repositories of documented subadult material, forcing Australian practitioners to rely on North American, arguably antiquated reference standards for age estimation. This study assessed the temporal pattern of ossification of the cranium and provides recalibrated probabilistic information for age estimation of modern Australian children. Fusion status of the occipital and frontal bones, atlas, and axis was scored using a modified two- to four-tier system from cranial/cervical DICOM datasets of 585 children aged birth to 10 years. Transition analysis was applied to elucidate maximum-likelihood estimates between consecutive fusion stages, in conjunction with Bayesian statistics to calculate credible intervals for age estimation. Results demonstrate significant sex differences in skeletal maturation (p < 0.05) and earlier timings in comparison with major literary sources, underscoring the requisite of updated standards for age estimation of modern individuals. © 2015 American Academy of Forensic Sciences.

Accuracy of diagnostic tests to detect asymptomatic bacteriuria during pregnancy.

PubMed

Mignini, Luciano; Carroli, Guillermo; Abalos, Edgardo; Widmer, Mariana; Amigot, Susana; Nardin, Juan Manuel; Giordano, Daniel; Merialdi, Mario; Arciero, Graciela; Del Carmen Hourquescos, Maria

2009-02-01

A dipslide is a plastic paddle coated with agar that is attached to a plastic cap that screws onto a sterile plastic vial. Our objective was to estimate the diagnostic accuracy of the dipslide culture technique to detect asymptomatic bacteriuria during pregnancy and to evaluate the accuracy of nitrate and leucocyte esterase dipslides for screening. This was an ancillary study within a trial comparing single-day with 7-day therapy in treating asymptomatic bacteriuria. Clean-catch midstream samples were collected from pregnant women seeking routine care. Positive and negative likelihood ratios and sensitivity and specificity for the culture-based dipslide to detect and chemical dipsticks (nitrites, leukocyte esterase, or both) to screen were estimated using traditional urine culture as the "gold standard." : A total of 3,048 eligible pregnant women were screened. The prevalence of asymptomatic bacteriuria was 15%, with Escherichia coli the most prevalent organism. The likelihood ratio for detecting asymptomatic bacteriuria with a positive dipslide test was 225 (95% confidence interval [CI] 113-449), increasing the probability of asymptomatic bacteriuria to 98%; the likelihood ratio for a negative dipslide test was 0.02 (95% CI 0.01-0.05), reducing the probability of bacteriuria to less than 1%. The positive likelihood ratio of leukocyte esterase and nitrite dipsticks (when both or either one was positive) was 6.95 (95% CI 5.80-8.33), increasing the probability of bacteriuria to only 54%; the negative likelihood ratio was 0.50 (95% CI 0.45-0.57), reducing the probability to 8%. A pregnant woman with a positive dipslide test is very likely to have a definitive diagnosis of asymptomatic bacteriuria, whereas a negative result effectively rules out the presence of bacteriuria. Dipsticks that measure nitrites and leukocyte esterase have low sensitivity for use in screening for asymptomatic bacteriuria during gestation. ISRCTN, isrctn.org, 1196608 II.

A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer.

PubMed

Paik, Soonmyung; Shak, Steven; Tang, Gong; Kim, Chungyeul; Baker, Joffre; Cronin, Maureen; Baehner, Frederick L; Walker, Michael G; Watson, Drew; Park, Taesung; Hiller, William; Fisher, Edwin R; Wickerham, D Lawrence; Bryant, John; Wolmark, Norman

2004-12-30

The likelihood of distant recurrence in patients with breast cancer who have no involved lymph nodes and estrogen-receptor-positive tumors is poorly defined by clinical and histopathological measures. We tested whether the results of a reverse-transcriptase-polymerase-chain-reaction (RT-PCR) assay of 21 prospectively selected genes in paraffin-embedded tumor tissue would correlate with the likelihood of distant recurrence in patients with node-negative, tamoxifen-treated breast cancer who were enrolled in the National Surgical Adjuvant Breast and Bowel Project clinical trial B-14. The levels of expression of 16 cancer-related genes and 5 reference genes were used in a prospectively defined algorithm to calculate a recurrence score and to determine a risk group (low, intermediate, or high) for each patient. Adequate RT-PCR profiles were obtained in 668 of 675 tumor blocks. The proportions of patients categorized as having a low, intermediate, or high risk by the RT-PCR assay were 51, 22, and 27 percent, respectively. The Kaplan-Meier estimates of the rates of distant recurrence at 10 years in the low-risk, intermediate-risk, and high-risk groups were 6.8 percent (95 percent confidence interval, 4.0 to 9.6), 14.3 percent (95 percent confidence interval, 8.3 to 20.3), and 30.5 percent (95 percent confidence interval, 23.6 to 37.4). The rate in the low-risk group was significantly lower than that in the high-risk group (P<0.001). In a multivariate Cox model, the recurrence score provided significant predictive power that was independent of age and tumor size (P<0.001). The recurrence score was also predictive of overall survival (P<0.001) and could be used as a continuous function to predict distant recurrence in individual patients. The recurrence score has been validated as quantifying the likelihood of distant recurrence in tamoxifen-treated patients with node-negative, estrogen-receptor-positive breast cancer. Copyright 2004 Massachusetts Medical Society.

Two new methods to fit models for network meta-analysis with random inconsistency effects.

PubMed

Law, Martin; Jackson, Dan; Turner, Rebecca; Rhodes, Kirsty; Viechtbauer, Wolfgang

2016-07-28

Meta-analysis is a valuable tool for combining evidence from multiple studies. Network meta-analysis is becoming more widely used as a means to compare multiple treatments in the same analysis. However, a network meta-analysis may exhibit inconsistency, whereby the treatment effect estimates do not agree across all trial designs, even after taking between-study heterogeneity into account. We propose two new estimation methods for network meta-analysis models with random inconsistency effects. The model we consider is an extension of the conventional random-effects model for meta-analysis to the network meta-analysis setting and allows for potential inconsistency using random inconsistency effects. Our first new estimation method uses a Bayesian framework with empirically-based prior distributions for both the heterogeneity and the inconsistency variances. We fit the model using importance sampling and thereby avoid some of the difficulties that might be associated with using Markov Chain Monte Carlo (MCMC). However, we confirm the accuracy of our importance sampling method by comparing the results to those obtained using MCMC as the gold standard. The second new estimation method we describe uses a likelihood-based approach, implemented in the metafor package, which can be used to obtain (restricted) maximum-likelihood estimates of the model parameters and profile likelihood confidence intervals of the variance components. We illustrate the application of the methods using two contrasting examples. The first uses all-cause mortality as an outcome, and shows little evidence of between-study heterogeneity or inconsistency. The second uses "ear discharge" as an outcome, and exhibits substantial between-study heterogeneity and inconsistency. Both new estimation methods give results similar to those obtained using MCMC. The extent of heterogeneity and inconsistency should be assessed and reported in any network meta-analysis. Our two new methods can be used to fit models for network meta-analysis with random inconsistency effects. They are easily implemented using the accompanying R code in the Additional file 1. Using these estimation methods, the extent of inconsistency can be assessed and reported.

A removal model for estimating detection probabilities from point-count surveys

USGS Publications Warehouse

Farnsworth, G.L.; Pollock, K.H.; Nichols, J.D.; Simons, T.R.; Hines, J.E.; Sauer, J.R.

2000-01-01

We adapted a removal model to estimate detection probability during point count surveys. The model assumes one factor influencing detection during point counts is the singing frequency of birds. This may be true for surveys recording forest songbirds when most detections are by sound. The model requires counts to be divided into several time intervals. We used time intervals of 2, 5, and 10 min to develop a maximum-likelihood estimator for the detectability of birds during such surveys. We applied this technique to data from bird surveys conducted in Great Smoky Mountains National Park. We used model selection criteria to identify whether detection probabilities varied among species, throughout the morning, throughout the season, and among different observers. The overall detection probability for all birds was 75%. We found differences in detection probability among species. Species that sing frequently such as Winter Wren and Acadian Flycatcher had high detection probabilities (about 90%) and species that call infrequently such as Pileated Woodpecker had low detection probability (36%). We also found detection probabilities varied with the time of day for some species (e.g. thrushes) and between observers for other species. This method of estimating detectability during point count surveys offers a promising new approach to using count data to address questions of the bird abundance, density, and population trends.

Variance Difference between Maximum Likelihood Estimation Method and Expected A Posteriori Estimation Method Viewed from Number of Test Items

ERIC Educational Resources Information Center

Mahmud, Jumailiyah; Sutikno, Muzayanah; Naga, Dali S.

2016-01-01

The aim of this study is to determine variance difference between maximum likelihood and expected A posteriori estimation methods viewed from number of test items of aptitude test. The variance presents an accuracy generated by both maximum likelihood and Bayes estimation methods. The test consists of three subtests, each with 40 multiple-choice…

A Comparison of a Bayesian and a Maximum Likelihood Tailored Testing Procedure.

ERIC Educational Resources Information Center

McKinley, Robert L.; Reckase, Mark D.

A study was conducted to compare tailored testing procedures based on a Bayesian ability estimation technique and on a maximum likelihood ability estimation technique. The Bayesian tailored testing procedure selected items so as to minimize the posterior variance of the ability estimate distribution, while the maximum likelihood tailored testing…

On non-parametric maximum likelihood estimation of the bivariate survivor function.

PubMed

Prentice, R L

The likelihood function for the bivariate survivor function F, under independent censorship, is maximized to obtain a non-parametric maximum likelihood estimator &Fcirc;. &Fcirc; may or may not be unique depending on the configuration of singly- and doubly-censored pairs. The likelihood function can be maximized by placing all mass on the grid formed by the uncensored failure times, or half lines beyond the failure time grid, or in the upper right quadrant beyond the grid. By accumulating the mass along lines (or regions) where the likelihood is flat, one obtains a partially maximized likelihood as a function of parameters that can be uniquely estimated. The score equations corresponding to these point mass parameters are derived, using a Lagrange multiplier technique to ensure unit total mass, and a modified Newton procedure is used to calculate the parameter estimates in some limited simulation studies. Some considerations for the further development of non-parametric bivariate survivor function estimators are briefly described.

Primary health care providers' advice for a dental checkup and dental use in children.

PubMed

Beil, Heather A; Rozier, R Gary

2010-08-01

In this study we estimated factors associated with children being advised to see the dentist by a doctor or other health provider; tested for an association between the advisement on the likelihood that the child would visit the dentist; and estimated the effect of the advisement on dental costs. We identified a sample of 5268 children aged 2 to 11 years in the 2004 Medical Expenditures Panel Survey. A cross-sectional analysis with logistic regression models was conducted to estimate the likelihood of the child receiving a recommendation for a dental checkup, and to determine its effect on the likelihood of having a dental visit. Differences in cost for children who received a recommendation were assessed by using a linear regression model. All analyses were conducted separately on children aged 2 to 5 (n = 2031) and aged 6 to 11 (n = 3237) years. Forty-seven percent of 2- to 5-year-olds and 37% of 6- to 11-year-olds had been advised to see the dentist. Children aged 2 to 5 who received a recommendation were more likely to have a dental visit (odds ratio: 2.89 [95% confidence interval: 2.16-3.87]), but no difference was observed among older children. Advice had no effect on dental costs in either age group. Health providers' recommendation that pediatric patients visit the dentist was associated with an increase in dental visits among young children. Providers have the potential to play an important role in establishing a dental home for children at an early age. Future research should examine potential interventions to increase effective dental referrals by health providers.

Reliability estimation of a N- M-cold-standby redundancy system in a multicomponent stress-strength model with generalized half-logistic distribution

NASA Astrophysics Data System (ADS)

Liu, Yiming; Shi, Yimin; Bai, Xuchao; Zhan, Pei

2018-01-01

In this paper, we study the estimation for the reliability of a multicomponent system, named N- M-cold-standby redundancy system, based on progressive Type-II censoring sample. In the system, there are N subsystems consisting of M statistically independent distributed strength components, and only one of these subsystems works under the impact of stresses at a time and the others remain as standbys. Whenever the working subsystem fails, one from the standbys takes its place. The system fails when the entire subsystems fail. It is supposed that the underlying distributions of random strength and stress both belong to the generalized half-logistic distribution with different shape parameter. The reliability of the system is estimated by using both classical and Bayesian statistical inference. Uniformly minimum variance unbiased estimator and maximum likelihood estimator for the reliability of the system are derived. Under squared error loss function, the exact expression of the Bayes estimator for the reliability of the system is developed by using the Gauss hypergeometric function. The asymptotic confidence interval and corresponding coverage probabilities are derived based on both the Fisher and the observed information matrices. The approximate highest probability density credible interval is constructed by using Monte Carlo method. Monte Carlo simulations are performed to compare the performances of the proposed reliability estimators. A real data set is also analyzed for an illustration of the findings.

Semi-Markov models for interval censored transient cognitive states with back transitions and a competing risk

PubMed Central

Wei, Shaoceng; Kryscio, Richard J.

2015-01-01

Continuous-time multi-state stochastic processes are useful for modeling the flow of subjects from intact cognition to dementia with mild cognitive impairment and global impairment as intervening transient, cognitive states and death as a competing risk (Figure 1). Each subject's cognition is assessed periodically resulting in interval censoring for the cognitive states while death without dementia is not interval censored. Since back transitions among the transient states are possible, Markov chains are often applied to this type of panel data. In this manuscript we apply a Semi-Markov process in which we assume that the waiting times are Weibull distributed except for transitions from the baseline state, which are exponentially distributed and in which we assume no additional changes in cognition occur between two assessments. We implement a quasi-Monte Carlo (QMC) method to calculate the higher order integration needed for likelihood estimation. We apply our model to a real dataset, the Nun Study, a cohort of 461 participants. PMID:24821001

Semi-Markov models for interval censored transient cognitive states with back transitions and a competing risk.

PubMed

Wei, Shaoceng; Kryscio, Richard J

2016-12-01

Continuous-time multi-state stochastic processes are useful for modeling the flow of subjects from intact cognition to dementia with mild cognitive impairment and global impairment as intervening transient cognitive states and death as a competing risk. Each subject's cognition is assessed periodically resulting in interval censoring for the cognitive states while death without dementia is not interval censored. Since back transitions among the transient states are possible, Markov chains are often applied to this type of panel data. In this manuscript, we apply a semi-Markov process in which we assume that the waiting times are Weibull distributed except for transitions from the baseline state, which are exponentially distributed and in which we assume no additional changes in cognition occur between two assessments. We implement a quasi-Monte Carlo (QMC) method to calculate the higher order integration needed for likelihood estimation. We apply our model to a real dataset, the Nun Study, a cohort of 461 participants. © The Author(s) 2014.

Exposure to Traffic-related Air Pollution During Pregnancy and Term Low Birth Weight: Estimation of Causal Associations in a Semiparametric Model

PubMed Central

Padula, Amy M.; Mortimer, Kathleen; Hubbard, Alan; Lurmann, Frederick; Jerrett, Michael; Tager, Ira B.

2012-01-01

Traffic-related air pollution is recognized as an important contributor to health problems. Epidemiologic analyses suggest that prenatal exposure to traffic-related air pollutants may be associated with adverse birth outcomes; however, there is insufficient evidence to conclude that the relation is causal. The Study of Air Pollution, Genetics and Early Life Events comprises all births to women living in 4 counties in California's San Joaquin Valley during the years 2000–2006. The probability of low birth weight among full-term infants in the population was estimated using machine learning and targeted maximum likelihood estimation for each quartile of traffic exposure during pregnancy. If everyone lived near high-volume freeways (approximated as the fourth quartile of traffic density), the estimated probability of term low birth weight would be 2.27% (95% confidence interval: 2.16, 2.38) as compared with 2.02% (95% confidence interval: 1.90, 2.12) if everyone lived near smaller local roads (first quartile of traffic density). Assessment of potentially causal associations, in the absence of arbitrary model assumptions applied to the data, should result in relatively unbiased estimates. The current results support findings from previous studies that prenatal exposure to traffic-related air pollution may adversely affect birth weight among full-term infants. PMID:23045474

The recursive maximum likelihood proportion estimator: User's guide and test results

NASA Technical Reports Server (NTRS)

Vanrooy, D. L.

1976-01-01

Implementation of the recursive maximum likelihood proportion estimator is described. A user's guide to programs as they currently exist on the IBM 360/67 at LARS, Purdue is included, and test results on LANDSAT data are described. On Hill County data, the algorithm yields results comparable to the standard maximum likelihood proportion estimator.

A semiparametric Bayesian proportional hazards model for interval censored data with frailty effects.

PubMed

Henschel, Volkmar; Engel, Jutta; Hölzel, Dieter; Mansmann, Ulrich

2009-02-10

Multivariate analysis of interval censored event data based on classical likelihood methods is notoriously cumbersome. Likelihood inference for models which additionally include random effects are not available at all. Developed algorithms bear problems for practical users like: matrix inversion, slow convergence, no assessment of statistical uncertainty. MCMC procedures combined with imputation are used to implement hierarchical models for interval censored data within a Bayesian framework. Two examples from clinical practice demonstrate the handling of clustered interval censored event times as well as multilayer random effects for inter-institutional quality assessment. The software developed is called survBayes and is freely available at CRAN. The proposed software supports the solution of complex analyses in many fields of clinical epidemiology as well as health services research.

The numerical evaluation of maximum-likelihood estimates of the parameters for a mixture of normal distributions from partially identified samples

NASA Technical Reports Server (NTRS)

Walker, H. F.

1976-01-01

Likelihood equations determined by the two types of samples which are necessary conditions for a maximum-likelihood estimate are considered. These equations, suggest certain successive-approximations iterative procedures for obtaining maximum-likelihood estimates. These are generalized steepest ascent (deflected gradient) procedures. It is shown that, with probability 1 as N sub 0 approaches infinity (regardless of the relative sizes of N sub 0 and N sub 1, i=1,...,m), these procedures converge locally to the strongly consistent maximum-likelihood estimates whenever the step size is between 0 and 2. Furthermore, the value of the step size which yields optimal local convergence rates is bounded from below by a number which always lies between 1 and 2.

A Power-Law Growth and Decay Model with Autocorrelation for Posting Data to Social Networking Services.

PubMed

Fujiyama, Toshifumi; Matsui, Chihiro; Takemura, Akimichi

2016-01-01

We propose a power-law growth and decay model for posting data to social networking services before and after social events. We model the time series structure of deviations from the power-law growth and decay with a conditional Poisson autoregressive (AR) model. Online postings related to social events are described by five parameters in the power-law growth and decay model, each of which characterizes different aspects of interest in the event. We assess the validity of parameter estimates in terms of confidence intervals, and compare various submodels based on likelihoods and information criteria.

Likelihood of early detection of breast cancer in relation to false-positive risk in life-time mammographic screening: population-based cohort study.

PubMed

Otten, J D M; Fracheboud, J; den Heeten, G J; Otto, S J; Holland, R; de Koning, H J; Broeders, M J M; Verbeek, A L M

2013-10-01

Women require balanced, high-quality information when making an informed decision on screening benefits and harms before attending biennial mammographic screening. The cumulative risk of a false-positive recall and/or (small) screen-detected or interval cancer over 13 consecutive screening examinations for women aged 50 from the start of screening were estimated using data from the Nijmegen programme, the Netherlands. Women who underwent 13 successive screens in the period 1975-1976 had a 5.3% cumulative chance of a screen-detected cancer, with a 4.2% risk of at least one false-positive recall. The risk of being diagnosed with interval cancer was 3.7%. Two decades later, these estimates were 6.9%, 7.3% and 2.9%, respectively. The chance of detection of a small, favourable invasive breast cancer, anticipating a normal life-expectancy, rose from 2.3% to 3.7%. Extrapolation to digital screening mammography indicates that the proportion of false-positive results will rise to 16%. Dutch women about to participate in the screening programme can be reassured that the chance of false-positive recall in the Netherlands is relatively low. A new screening policy and improved mammography have increased the detection of an early screening carcinoma and lowering the risk of interval carcinoma.

Evaluation of some random effects methodology applicable to bird ringing data

USGS Publications Warehouse

Burnham, K.P.; White, Gary C.

2002-01-01

Existing models for ring recovery and recapture data analysis treat temporal variations in annual survival probability (S) as fixed effects. Often there is no explainable structure to the temporal variation in S1,..., Sk; random effects can then be a useful model: Si = E(S) + ??i. Here, the temporal variation in survival probability is treated as random with average value E(??2) = ??2. This random effects model can now be fit in program MARK. Resultant inferences include point and interval estimation for process variation, ??2, estimation of E(S) and var (E??(S)) where the latter includes a component for ??2 as well as the traditional component for v??ar(S??\\S??). Furthermore, the random effects model leads to shrinkage estimates, Si, as improved (in mean square error) estimators of Si compared to the MLE, S??i, from the unrestricted time-effects model. Appropriate confidence intervals based on the Si are also provided. In addition, AIC has been generalized to random effects models. This paper presents results of a Monte Carlo evaluation of inference performance under the simple random effects model. Examined by simulation, under the simple one group Cormack-Jolly-Seber (CJS) model, are issues such as bias of ??s2, confidence interval coverage on ??2, coverage and mean square error comparisons for inference about Si based on shrinkage versus maximum likelihood estimators, and performance of AIC model selection over three models: Si ??? S (no effects), Si = E(S) + ??i (random effects), and S1,..., Sk (fixed effects). For the cases simulated, the random effects methods performed well and were uniformly better than fixed effects MLE for the Si.

Mixture models for undiagnosed prevalent disease and interval-censored incident disease: applications to a cohort assembled from electronic health records.

PubMed

Cheung, Li C; Pan, Qing; Hyun, Noorie; Schiffman, Mark; Fetterman, Barbara; Castle, Philip E; Lorey, Thomas; Katki, Hormuzd A

2017-09-30

For cost-effectiveness and efficiency, many large-scale general-purpose cohort studies are being assembled within large health-care providers who use electronic health records. Two key features of such data are that incident disease is interval-censored between irregular visits and there can be pre-existing (prevalent) disease. Because prevalent disease is not always immediately diagnosed, some disease diagnosed at later visits are actually undiagnosed prevalent disease. We consider prevalent disease as a point mass at time zero for clinical applications where there is no interest in time of prevalent disease onset. We demonstrate that the naive Kaplan-Meier cumulative risk estimator underestimates risks at early time points and overestimates later risks. We propose a general family of mixture models for undiagnosed prevalent disease and interval-censored incident disease that we call prevalence-incidence models. Parameters for parametric prevalence-incidence models, such as the logistic regression and Weibull survival (logistic-Weibull) model, are estimated by direct likelihood maximization or by EM algorithm. Non-parametric methods are proposed to calculate cumulative risks for cases without covariates. We compare naive Kaplan-Meier, logistic-Weibull, and non-parametric estimates of cumulative risk in the cervical cancer screening program at Kaiser Permanente Northern California. Kaplan-Meier provided poor estimates while the logistic-Weibull model was a close fit to the non-parametric. Our findings support our use of logistic-Weibull models to develop the risk estimates that underlie current US risk-based cervical cancer screening guidelines. Published 2017. This article has been contributed to by US Government employees and their work is in the public domain in the USA. Published 2017. This article has been contributed to by US Government employees and their work is in the public domain in the USA.

Quasi- and pseudo-maximum likelihood estimators for discretely observed continuous-time Markov branching processes

PubMed Central

Chen, Rui; Hyrien, Ollivier

2011-01-01

This article deals with quasi- and pseudo-likelihood estimation in a class of continuous-time multi-type Markov branching processes observed at discrete points in time. “Conventional” and conditional estimation are discussed for both approaches. We compare their properties and identify situations where they lead to asymptotically equivalent estimators. Both approaches possess robustness properties, and coincide with maximum likelihood estimation in some cases. Quasi-likelihood functions involving only linear combinations of the data may be unable to estimate all model parameters. Remedial measures exist, including the resort either to non-linear functions of the data or to conditioning the moments on appropriate sigma-algebras. The method of pseudo-likelihood may also resolve this issue. We investigate the properties of these approaches in three examples: the pure birth process, the linear birth-and-death process, and a two-type process that generalizes the previous two examples. Simulations studies are conducted to evaluate performance in finite samples. PMID:21552356

Spatial hydrological drought characteristics in Karkheh River basin, southwest Iran using copulas

NASA Astrophysics Data System (ADS)

Dodangeh, Esmaeel; Shahedi, Kaka; Shiau, Jenq-Tzong; MirAkbari, Maryam

2017-08-01

Investigation on drought characteristics such as severity, duration, and frequency is crucial for water resources planning and management in a river basin. While the methodology for multivariate drought frequency analysis is well established by applying the copulas, the estimation on the associated parameters by various parameter estimation methods and the effects on the obtained results have not yet been investigated. This research aims at conducting a comparative analysis between the maximum likelihood parametric and non-parametric method of the Kendall τ estimation method for copulas parameter estimation. The methods were employed to study joint severity-duration probability and recurrence intervals in Karkheh River basin (southwest Iran) which is facing severe water-deficit problems. Daily streamflow data at three hydrological gauging stations (Tang Sazbon, Huleilan and Polchehr) near the Karkheh dam were used to draw flow duration curves (FDC) of these three stations. The Q_{75} index extracted from the FDC were set as threshold level to abstract drought characteristics such as drought duration and severity on the basis of the run theory. Drought duration and severity were separately modeled using the univariate probabilistic distributions and gamma-GEV, LN2-exponential, and LN2-gamma were selected as the best paired drought severity-duration inputs for copulas according to the Akaike Information Criteria (AIC), Kolmogorov-Smirnov and chi-square tests. Archimedean Clayton, Frank, and extreme value Gumbel copulas were employed to construct joint cumulative distribution functions (JCDF) of droughts for each station. Frank copula at Tang Sazbon and Gumbel at Huleilan and Polchehr stations were identified as the best copulas based on the performance evaluation criteria including AIC, BIC, log-likelihood and root mean square error (RMSE) values. Based on the RMSE values, nonparametric Kendall-τ is preferred to the parametric maximum likelihood estimation method. The results showed greater drought return periods by the parametric ML method in comparison to the nonparametric Kendall τ estimation method. The results also showed that stations located in tributaries (Huleilan and Polchehr) have close return periods, while the station along the main river (Tang Sazbon) has the smaller return periods for the drought events with identical drought duration and severity.

Bias Correction for the Maximum Likelihood Estimate of Ability. Research Report. ETS RR-05-15

ERIC Educational Resources Information Center

Zhang, Jinming

2005-01-01

Lord's bias function and the weighted likelihood estimation method are effective in reducing the bias of the maximum likelihood estimate of an examinee's ability under the assumption that the true item parameters are known. This paper presents simulation studies to determine the effectiveness of these two methods in reducing the bias when the item…

Fast maximum likelihood estimation of mutation rates using a birth-death process.

PubMed

Wu, Xiaowei; Zhu, Hongxiao

2015-02-07

Since fluctuation analysis was first introduced by Luria and Delbrück in 1943, it has been widely used to make inference about spontaneous mutation rates in cultured cells. Under certain model assumptions, the probability distribution of the number of mutants that appear in a fluctuation experiment can be derived explicitly, which provides the basis of mutation rate estimation. It has been shown that, among various existing estimators, the maximum likelihood estimator usually demonstrates some desirable properties such as consistency and lower mean squared error. However, its application in real experimental data is often hindered by slow computation of likelihood due to the recursive form of the mutant-count distribution. We propose a fast maximum likelihood estimator of mutation rates, MLE-BD, based on a birth-death process model with non-differential growth assumption. Simulation studies demonstrate that, compared with the conventional maximum likelihood estimator derived from the Luria-Delbrück distribution, MLE-BD achieves substantial improvement on computational speed and is applicable to arbitrarily large number of mutants. In addition, it still retains good accuracy on point estimation. Published by Elsevier Ltd.

Autonomous mechanism of internal choice estimate underlies decision inertia.

PubMed

Akaishi, Rei; Umeda, Kazumasa; Nagase, Asako; Sakai, Katsuyuki

2014-01-08

Our choice is influenced by choices we made in the past, but the mechanism responsible for the choice bias remains elusive. Here we show that the history-dependent choice bias can be explained by an autonomous learning rule whereby an estimate of the likelihood of a choice to be made is updated in each trial by comparing between the actual and expected choices. We found that in perceptual decision making without performance feedback, a decision on an ambiguous stimulus is repeated on the subsequent trial more often than a decision on a salient stimulus. This inertia of decision was not accounted for by biases in motor response, sensory processing, or attention. The posterior cingulate cortex and frontal eye field represent choice prediction error and choice estimate in the learning algorithm, respectively. Interactions between the two regions during the intertrial interval are associated with decision inertia on a subsequent trial. Copyright © 2014 Elsevier Inc. All rights reserved.

Spatial Distribution of the Coefficient of Variation for the Paleo-Earthquakes in Japan

NASA Astrophysics Data System (ADS)

Nomura, S.; Ogata, Y.

2015-12-01

Renewal processes, point prccesses in which intervals between consecutive events are independently and identically distributed, are frequently used to describe this repeating earthquake mechanism and forecast the next earthquakes. However, one of the difficulties in applying recurrent earthquake models is the scarcity of the historical data. Most studied fault segments have few, or only one observed earthquake that often have poorly constrained historic and/or radiocarbon ages. The maximum likelihood estimate from such a small data set can have a large bias and error, which tends to yield high probability for the next event in a very short time span when the recurrence intervals have similar lengths. On the other hand, recurrence intervals at a fault depend on the long-term slip rate caused by the tectonic motion in average. In addition, recurrence times are also fluctuated by nearby earthquakes or fault activities which encourage or discourage surrounding seismicity. These factors have spatial trends due to the heterogeneity of tectonic motion and seismicity. Thus, this paper introduces a spatial structure on the key parameters of renewal processes for recurrent earthquakes and estimates it by using spatial statistics. Spatial variation of mean and variance parameters of recurrence times are estimated in Bayesian framework and the next earthquakes are forecasted by Bayesian predictive distributions. The proposal model is applied for recurrent earthquake catalog in Japan and its result is compared with the current forecast adopted by the Earthquake Research Committee of Japan.

Estimation of the ARNO model baseflow parameters using daily streamflow data

NASA Astrophysics Data System (ADS)

Abdulla, F. A.; Lettenmaier, D. P.; Liang, Xu

1999-09-01

An approach is described for estimation of baseflow parameters of the ARNO model, using historical baseflow recession sequences extracted from daily streamflow records. This approach allows four of the model parameters to be estimated without rainfall data, and effectively facilitates partitioning of the parameter estimation procedure so that parsimonious search procedures can be used to estimate the remaining storm response parameters separately. Three methods of optimization are evaluated for estimation of four baseflow parameters. These methods are the downhill Simplex (S), Simulated Annealing combined with the Simplex method (SA) and Shuffled Complex Evolution (SCE). These estimation procedures are explored in conjunction with four objective functions: (1) ordinary least squares; (2) ordinary least squares with Box-Cox transformation; (3) ordinary least squares on prewhitened residuals; (4) ordinary least squares applied to prewhitened with Box-Cox transformation of residuals. The effects of changing the seed random generator for both SA and SCE methods are also explored, as are the effects of the bounds of the parameters. Although all schemes converge to the same values of the objective function, SCE method was found to be less sensitive to these issues than both the SA and the Simplex schemes. Parameter uncertainty and interactions are investigated through estimation of the variance-covariance matrix and confidence intervals. As expected the parameters were found to be correlated and the covariance matrix was found to be not diagonal. Furthermore, the linearized confidence interval theory failed for about one-fourth of the catchments while the maximum likelihood theory did not fail for any of the catchments.

An alternative method to measure the likelihood of a financial crisis in an emerging market

NASA Astrophysics Data System (ADS)

Özlale, Ümit; Metin-Özcan, Kıvılcım

2007-07-01

This paper utilizes an early warning system in order to measure the likelihood of a financial crisis in an emerging market economy. We introduce a methodology, where we can both obtain a likelihood series and analyze the time-varying effects of several macroeconomic variables on this likelihood. Since the issue is analyzed in a non-linear state space framework, the extended Kalman filter emerges as the optimal estimation algorithm. Taking the Turkish economy as our laboratory, the results indicate that both the derived likelihood measure and the estimated time-varying parameters are meaningful and can successfully explain the path that the Turkish economy had followed between 2000 and 2006. The estimated parameters also suggest that overvalued domestic currency, current account deficit and the increase in the default risk increase the likelihood of having an economic crisis in the economy. Overall, the findings in this paper suggest that the estimation methodology introduced in this paper can also be applied to other emerging market economies as well.

Maximum likelihood estimation of signal-to-noise ratio and combiner weight

NASA Technical Reports Server (NTRS)

Kalson, S.; Dolinar, S. J.

1986-01-01

An algorithm for estimating signal to noise ratio and combiner weight parameters for a discrete time series is presented. The algorithm is based upon the joint maximum likelihood estimate of the signal and noise power. The discrete-time series are the sufficient statistics obtained after matched filtering of a biphase modulated signal in additive white Gaussian noise, before maximum likelihood decoding is performed.

Comparison of Maximum Likelihood Estimation Approach and Regression Approach in Detecting Quantitative Trait Lco Using RAPD Markers

Treesearch

Changren Weng; Thomas L. Kubisiak; C. Dana Nelson; James P. Geaghan; Michael Stine

1999-01-01

Single marker regression and single marker maximum likelihood estimation were tied to detect quantitative trait loci (QTLs) controlling the early height growth of longleaf pine and slash pine using a ((longleaf pine x slash pine) x slash pine) BC, population consisting of 83 progeny. Maximum likelihood estimation was found to be more power than regression and could...

Nonparametric spirometry reference values for Hispanic Americans.

PubMed

Glenn, Nancy L; Brown, Vanessa M

2011-02-01

Recent literature sites ethnic origin as a major factor in developing pulmonary function reference values. Extensive studies established reference values for European and African Americans, but not for Hispanic Americans. The Third National Health and Nutrition Examination Survey defines Hispanic as individuals of Spanish speaking cultures. While no group was excluded from the target population, sample size requirements only allowed inclusion of individuals who identified themselves as Mexican Americans. This research constructs nonparametric reference value confidence intervals for Hispanic American pulmonary function. The method is applicable to all ethnicities. We use empirical likelihood confidence intervals to establish normal ranges for reference values. Its major advantage: it is model free, but shares asymptotic properties of model based methods. Statistical comparisons indicate that empirical likelihood interval lengths are comparable to normal theory intervals. Power and efficiency studies agree with previously published theoretical results.

Determinants of attaining and maintaining a low cardiovascular risk profile--the Doetinchem Cohort Study.

PubMed

Hulsegge, Gerben; van der Schouw, Yvonne T; Daviglus, Martha L; Smit, Henriëtte A; Verschuren, W M Monique

2016-02-01

While maintenance of a low cardiovascular risk profile is essential for cardiovascular disease (CVD) prevention, few people maintain a low CVD risk profile throughout their life. We studied the association of demographic, lifestyle, psychological factors and family history of CVD with attainment and maintenance of a low risk profile over three subsequent 5-year periods. Measurements of 6390 adults aged 26-65 years at baseline were completed from 1993 to 97 and subsequently at 5-year intervals until 2013. At each wave, participants were categorized into low risk profile (ideal levels of blood pressure, cholesterol and body mass index, non-smoking and no diabetes) and medium/high risk profile (all others). Multivariable-adjusted modified Poisson regression analyses were used to examine determinants of attainment and maintenance of low risk; risk ratios (RR) and 95% confidence intervals (95% CI) were obtained. Generalized estimating equations were used to combine multiple 5-year comparisons. Younger age, female gender and high educational level were associated with higher likelihood of both maintaining and attaining low risk profile (P < 0.05). In addition, likelihood of attaining low risk was 9% higher with each 1-unit increment in Mediterranean diet score (RR: 1.09, 95% CI: 1.02-1.16), twice as high with any physical activity versus none (RR: 2.17, 95% CI: 1.16-4.04) and 35% higher with moderate alcohol consumption versus heavy consumption (RR: 1.35, 95% CI: 1.06-1.73). Healthy lifestyle factors such as adherence to a Mediterranean diet, physical activity and moderate as opposed to heavy alcohol consumption were associated with a higher likelihood of attaining a low risk profile. © The Author 2015. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.

Investigating the Impact of Uncertainty about Item Parameters on Ability Estimation

ERIC Educational Resources Information Center

Zhang, Jinming; Xie, Minge; Song, Xiaolan; Lu, Ting

2011-01-01

Asymptotic expansions of the maximum likelihood estimator (MLE) and weighted likelihood estimator (WLE) of an examinee's ability are derived while item parameter estimators are treated as covariates measured with error. The asymptotic formulae present the amount of bias of the ability estimators due to the uncertainty of item parameter estimators.…

Accuracy of autotitrating CPAP to estimate the residual Apnea-Hypopnea Index in patients with obstructive sleep apnea on treatment with autotitrating CPAP.

PubMed

Desai, Himanshu; Patel, Anil; Patel, Pinal; Grant, Brydon J B; Mador, M Jeffery

2009-11-01

Autotitrating continuous positive airway pressure (auto-CPAP) devices now have a smart card (a pocket-sized card with embedded integrated circuits which records data from the CPAP machine such as CPAP usage, CPAP pressure, large leak, etc.) which can estimate the Apnea-Hypopnea Index (AHI) on therapy. The aim of this study was to determine the accuracy of auto-CPAP in estimating the residual AHI in patients with obstructive sleep apnea (OSA) who were treated with auto-CPAP without a CPAP titration study. We studied 99 patients with OSA from April 2005 to May 2007 who underwent a repeat sleep study using auto-CPAP. The estimated AHI from auto-CPAP was compared with the AHI from an overnight polysomnogram (PSG) on auto-CPAP using Bland-Altman plot and likelihood ratio analyses. A PSG AHI cutoff of five events per hour was used to differentiate patients optimally treated with auto-CPAP from those with residual OSA on therapy. Bland and Altman analysis showed good agreement between auto-CPAP AHI and PSG AHI. There was no significant bias when smart card estimates of AHI at home were compared to smart card estimates obtained in the sleep laboratory. An auto-CPAP cutoff for the AHI of six events per hour was shown to be optimal for differentiating patients with and without residual OSA with a sensitivity of 0.92 (95% confidence interval (CI) 0.76 to 0.98) and specificity of 0.90 (95% CI 0.82 to 0.95) with a positive likelihood ratio (LR) of 9.6 (95% CI 5.1 to 21.5) and a negative likelihood ratio of 0.085 (95% CI 0.02 to 0.25). Auto-CPAP AHI of eight events per hour yielded the optimal sensitivity (0.94, 95% CI 0.73 to 0.99) and specificity (0.90, 95% CI 0.82 to 0.95) with a positive LR of 9.6 (95% CI 5.23 to 20.31) and a negative LR of 0.065 (95% CI 0.004 to 0.279) to identify patients with a PSG AHI of > or = 10 events per hour. Auto-CPAP estimate of AHI may be used to estimate residual AHI in patients with OSA of varying severity treated with auto-CPAP.

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

Shaffer, Richard, E-mail: rickyshaffer@yahoo.co.u; Department of Clinical Oncology, Imperial College London National Health Service Trust, London; Pickles, Tom

Purpose: Prior studies have derived low values of alpha-beta ratio (a/ss) for prostate cancer of approximately 1-2 Gy. These studies used poorly matched groups, differing definitions of biochemical failure, and insufficient follow-up. Methods and Materials: National Comprehensive Cancer Network low- or low-intermediate risk prostate cancer patients, treated with external beam radiotherapy or permanent prostate brachytherapy, were matched for prostate-specific antigen, Gleason score, T-stage, percentage of positive cores, androgen deprivation therapy, and era, yielding 118 patient pairs. The Phoenix definition of biochemical failure was used. The best-fitting value for a/ss was found for up to 90-month follow-up using maximum likelihood analysis,more » and the 95% confidence interval using the profile likelihood method. Linear quadratic formalism was applied with the radiobiological parameters of relative biological effectiveness = 1.0, potential doubling time = 45 days, and repair half-time = 1 hour. Bootstrap analysis was performed to estimate uncertainties in outcomes, and hence in a/ss. Sensitivity analysis was performed by varying the values of the radiobiological parameters to extreme values. Results: The value of a/ss best fitting the outcomes data was >30 Gy, with lower 95% confidence limit of 5.2 Gy. This was confirmed on bootstrap analysis. Varying parameters to extreme values still yielded best-fit a/ss of >30 Gy, although the lower 95% confidence interval limit was reduced to 0.6 Gy. Conclusions: Using carefully matched groups, long follow-up, the Phoenix definition of biochemical failure, and well-established statistical methods, the best estimate of a/ss for low and low-tier intermediate-risk prostate cancer is likely to be higher than that of normal tissues, although a low value cannot be excluded.« less

Maximum likelihood estimation of finite mixture model for economic data

NASA Astrophysics Data System (ADS)

Phoong, Seuk-Yen; Ismail, Mohd Tahir

2014-06-01

Finite mixture model is a mixture model with finite-dimension. This models are provides a natural representation of heterogeneity in a finite number of latent classes. In addition, finite mixture models also known as latent class models or unsupervised learning models. Recently, maximum likelihood estimation fitted finite mixture models has greatly drawn statistician's attention. The main reason is because maximum likelihood estimation is a powerful statistical method which provides consistent findings as the sample sizes increases to infinity. Thus, the application of maximum likelihood estimation is used to fit finite mixture model in the present paper in order to explore the relationship between nonlinear economic data. In this paper, a two-component normal mixture model is fitted by maximum likelihood estimation in order to investigate the relationship among stock market price and rubber price for sampled countries. Results described that there is a negative effect among rubber price and stock market price for Malaysia, Thailand, Philippines and Indonesia.

A Solution to Separation and Multicollinearity in Multiple Logistic Regression

PubMed Central

Shen, Jianzhao; Gao, Sujuan

2010-01-01

In dementia screening tests, item selection for shortening an existing screening test can be achieved using multiple logistic regression. However, maximum likelihood estimates for such logistic regression models often experience serious bias or even non-existence because of separation and multicollinearity problems resulting from a large number of highly correlated items. Firth (1993, Biometrika, 80(1), 27–38) proposed a penalized likelihood estimator for generalized linear models and it was shown to reduce bias and the non-existence problems. The ridge regression has been used in logistic regression to stabilize the estimates in cases of multicollinearity. However, neither solves the problems for each other. In this paper, we propose a double penalized maximum likelihood estimator combining Firth’s penalized likelihood equation with a ridge parameter. We present a simulation study evaluating the empirical performance of the double penalized likelihood estimator in small to moderate sample sizes. We demonstrate the proposed approach using a current screening data from a community-based dementia study. PMID:20376286

A Solution to Separation and Multicollinearity in Multiple Logistic Regression.

PubMed

Shen, Jianzhao; Gao, Sujuan

2008-10-01

In dementia screening tests, item selection for shortening an existing screening test can be achieved using multiple logistic regression. However, maximum likelihood estimates for such logistic regression models often experience serious bias or even non-existence because of separation and multicollinearity problems resulting from a large number of highly correlated items. Firth (1993, Biometrika, 80(1), 27-38) proposed a penalized likelihood estimator for generalized linear models and it was shown to reduce bias and the non-existence problems. The ridge regression has been used in logistic regression to stabilize the estimates in cases of multicollinearity. However, neither solves the problems for each other. In this paper, we propose a double penalized maximum likelihood estimator combining Firth's penalized likelihood equation with a ridge parameter. We present a simulation study evaluating the empirical performance of the double penalized likelihood estimator in small to moderate sample sizes. We demonstrate the proposed approach using a current screening data from a community-based dementia study.

Maximum Likelihood Estimation with Emphasis on Aircraft Flight Data

NASA Technical Reports Server (NTRS)

Iliff, K. W.; Maine, R. E.

1985-01-01

Accurate modeling of flexible space structures is an important field that is currently under investigation. Parameter estimation, using methods such as maximum likelihood, is one of the ways that the model can be improved. The maximum likelihood estimator has been used to extract stability and control derivatives from flight data for many years. Most of the literature on aircraft estimation concentrates on new developments and applications, assuming familiarity with basic estimation concepts. Some of these basic concepts are presented. The maximum likelihood estimator and the aircraft equations of motion that the estimator uses are briefly discussed. The basic concepts of minimization and estimation are examined for a simple computed aircraft example. The cost functions that are to be minimized during estimation are defined and discussed. Graphic representations of the cost functions are given to help illustrate the minimization process. Finally, the basic concepts are generalized, and estimation from flight data is discussed. Specific examples of estimation of structural dynamics are included. Some of the major conclusions for the computed example are also developed for the analysis of flight data.

An iterative procedure for obtaining maximum-likelihood estimates of the parameters for a mixture of normal distributions, Addendum

NASA Technical Reports Server (NTRS)

Peters, B. C., Jr.; Walker, H. F.

1975-01-01

New results and insights concerning a previously published iterative procedure for obtaining maximum-likelihood estimates of the parameters for a mixture of normal distributions were discussed. It was shown that the procedure converges locally to the consistent maximum likelihood estimate as long as a specified parameter is bounded between two limits. Bound values were given to yield optimal local convergence.

An iterative procedure for obtaining maximum-likelihood estimates of the parameters for a mixture of normal distributions

NASA Technical Reports Server (NTRS)

Peters, B. C., Jr.; Walker, H. F.

1975-01-01

A general iterative procedure is given for determining the consistent maximum likelihood estimates of normal distributions. In addition, a local maximum of the log-likelihood function, Newtons's method, a method of scoring, and modifications of these procedures are discussed.

SLDAssay: A software package and web tool for analyzing limiting dilution assays.

PubMed

Trumble, Ilana M; Allmon, Andrew G; Archin, Nancie M; Rigdon, Joseph; Francis, Owen; Baldoni, Pedro L; Hudgens, Michael G

2017-11-01

Serial limiting dilution (SLD) assays are used in many areas of infectious disease related research. This paper presents SLDAssay, a free and publicly available R software package and web tool for analyzing data from SLD assays. SLDAssay computes the maximum likelihood estimate (MLE) for the concentration of target cells, with corresponding exact and asymptotic confidence intervals. Exact and asymptotic goodness of fit p-values, and a bias-corrected (BC) MLE are also provided. No other publicly available software currently implements the BC MLE or the exact methods. For validation of SLDAssay, results from Myers et al. (1994) are replicated. Simulations demonstrate the BC MLE is less biased than the MLE. Additionally, simulations demonstrate that exact methods tend to give better confidence interval coverage and goodness-of-fit tests with lower type I error than the asymptotic methods. Additional advantages of using exact methods are also discussed. Copyright © 2017 Elsevier B.V. All rights reserved.

On the Relationships between Jeffreys Modal and Weighted Likelihood Estimation of Ability under Logistic IRT Models

ERIC Educational Resources Information Center

Magis, David; Raiche, Gilles

2012-01-01

This paper focuses on two estimators of ability with logistic item response theory models: the Bayesian modal (BM) estimator and the weighted likelihood (WL) estimator. For the BM estimator, Jeffreys' prior distribution is considered, and the corresponding estimator is referred to as the Jeffreys modal (JM) estimator. It is established that under…

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

Multiple-hit parameter estimation in monolithic detectors.

PubMed

Hunter, William C J; Barrett, Harrison H; Lewellen, Tom K; Miyaoka, Robert S

2013-02-01

We examine a maximum-a-posteriori method for estimating the primary interaction position of gamma rays with multiple interaction sites (hits) in a monolithic detector. In assessing the performance of a multiple-hit estimator over that of a conventional one-hit estimator, we consider a few different detector and readout configurations of a 50-mm-wide square cerium-doped lutetium oxyorthosilicate block. For this study, we use simulated data from SCOUT, a Monte-Carlo tool for photon tracking and modeling scintillation- camera output. With this tool, we determine estimate bias and variance for a multiple-hit estimator and compare these with similar metrics for a one-hit maximum-likelihood estimator, which assumes full energy deposition in one hit. We also examine the effect of event filtering on these metrics; for this purpose, we use a likelihood threshold to reject signals that are not likely to have been produced under the assumed likelihood model. Depending on detector design, we observe a 1%-12% improvement of intrinsic resolution for a 1-or-2-hit estimator as compared with a 1-hit estimator. We also observe improved differentiation of photopeak events using a 1-or-2-hit estimator as compared with the 1-hit estimator; more than 6% of photopeak events that were rejected by likelihood filtering for the 1-hit estimator were accurately identified as photopeak events and positioned without loss of resolution by a 1-or-2-hit estimator; for PET, this equates to at least a 12% improvement in coincidence-detection efficiency with likelihood filtering applied.

An evaluation of percentile and maximum likelihood estimators of weibull paremeters

Treesearch

Stanley J. Zarnoch; Tommy R. Dell

1985-01-01

Two methods of estimating the three-parameter Weibull distribution were evaluated by computer simulation and field data comparison. Maximum likelihood estimators (MLB) with bias correction were calculated with the computer routine FITTER (Bailey 1974); percentile estimators (PCT) were those proposed by Zanakis (1979). The MLB estimators had superior smaller bias and...

The Equivalence of Two Methods of Parameter Estimation for the Rasch Model.

ERIC Educational Resources Information Center

Blackwood, Larry G.; Bradley, Edwin L.

1989-01-01

Two methods of estimating parameters in the Rasch model are compared. The equivalence of likelihood estimations from the model of G. J. Mellenbergh and P. Vijn (1981) and from usual unconditional maximum likelihood (UML) estimation is demonstrated. Mellenbergh and Vijn's model is a convenient method of calculating UML estimates. (SLD)

Effects of Estimation Bias on Multiple-Category Classification with an IRT-Based Adaptive Classification Procedure

ERIC Educational Resources Information Center

Yang, Xiangdong; Poggio, John C.; Glasnapp, Douglas R.

2006-01-01

The effects of five ability estimators, that is, maximum likelihood estimator, weighted likelihood estimator, maximum a posteriori, expected a posteriori, and Owen's sequential estimator, on the performances of the item response theory-based adaptive classification procedure on multiple categories were studied via simulations. The following…

Essays in the California electricity reserves markets

NASA Astrophysics Data System (ADS)

Metaxoglou, Konstantinos

This dissertation examines inefficiencies in the California electricity reserves markets. In Chapter 1, I use the information released during the investigation of the state's electricity crisis of 2000 and 2001 by the Federal Energy Regulatory Commission to diagnose allocative inefficiencies. Building upon the work of Wolak (2000), I calculate a lower bound for the sellers' price-cost margins using the inverse elasticities of their residual demand curves. The downward bias in my estimates stems from the fact that I don't account for the hierarchical substitutability of the reserve types. The margins averaged at least 20 percent for the two highest quality types of reserves, regulation and spinning, generating millions of dollars in transfers to a handful of sellers. I provide evidence that the deviations from marginal cost pricing were due to the markets' high concentration and a principal-agent relationship that emerged from their design. In Chapter 2, I document systematic differences between the markets' day- and hour-ahead prices. I use a high-dimensional vector moving average model to estimate the premia and conduct correct inferences. To obtain exact maximum likelihood estimates of the model, I employ the EM algorithm that I develop in Chapter 3. I uncover significant day-ahead premia, which I attribute to market design characteristics too. On the demand side, the market design established a principal-agent relationship between the markets' buyers (principal) and their supervisory authority (agent). The agent had very limited incentives to shift reserve purchases to the lower priced hour-ahead markets. On the supply side, the market design raised substantial entry barriers by precluding purely speculative trading and by introducing a complicated code of conduct that induced uncertainty about which actions were subject to regulatory scrutiny. In Chapter 3, I introduce a state-space representation for vector autoregressive moving average models that enables exact maximum likelihood estimation using the EM algorithm. Moreover, my algorithm uses only analytical expressions; it requires the Kalman filter and a fixed-interval smoother in the E step and least squares-type regression in the M step. In contrast, existing maximum likelihood estimation methods require numerical differentiation, both for univariate and multivariate models.

SMURC: High-Dimension Small-Sample Multivariate Regression With Covariance Estimation.

PubMed

Bayar, Belhassen; Bouaynaya, Nidhal; Shterenberg, Roman

2017-03-01

We consider a high-dimension low sample-size multivariate regression problem that accounts for correlation of the response variables. The system is underdetermined as there are more parameters than samples. We show that the maximum likelihood approach with covariance estimation is senseless because the likelihood diverges. We subsequently propose a normalization of the likelihood function that guarantees convergence. We call this method small-sample multivariate regression with covariance (SMURC) estimation. We derive an optimization problem and its convex approximation to compute SMURC. Simulation results show that the proposed algorithm outperforms the regularized likelihood estimator with known covariance matrix and the sparse conditional Gaussian graphical model. We also apply SMURC to the inference of the wing-muscle gene network of the Drosophila melanogaster (fruit fly).

Maximum likelihood solution for inclination-only data in paleomagnetism

NASA Astrophysics Data System (ADS)

Arason, P.; Levi, S.

2010-08-01

We have developed a new robust maximum likelihood method for estimating the unbiased mean inclination from inclination-only data. In paleomagnetic analysis, the arithmetic mean of inclination-only data is known to introduce a shallowing bias. Several methods have been introduced to estimate the unbiased mean inclination of inclination-only data together with measures of the dispersion. Some inclination-only methods were designed to maximize the likelihood function of the marginal Fisher distribution. However, the exact analytical form of the maximum likelihood function is fairly complicated, and all the methods require various assumptions and approximations that are often inappropriate. For some steep and dispersed data sets, these methods provide estimates that are significantly displaced from the peak of the likelihood function to systematically shallower inclination. The problem locating the maximum of the likelihood function is partly due to difficulties in accurately evaluating the function for all values of interest, because some elements of the likelihood function increase exponentially as precision parameters increase, leading to numerical instabilities. In this study, we succeeded in analytically cancelling exponential elements from the log-likelihood function, and we are now able to calculate its value anywhere in the parameter space and for any inclination-only data set. Furthermore, we can now calculate the partial derivatives of the log-likelihood function with desired accuracy, and locate the maximum likelihood without the assumptions required by previous methods. To assess the reliability and accuracy of our method, we generated large numbers of random Fisher-distributed data sets, for which we calculated mean inclinations and precision parameters. The comparisons show that our new robust Arason-Levi maximum likelihood method is the most reliable, and the mean inclination estimates are the least biased towards shallow values.

On the log-normality of historical magnetic-storm intensity statistics: implications for extreme-event probabilities

USGS Publications Warehouse

Love, Jeffrey J.; Rigler, E. Joshua; Pulkkinen, Antti; Riley, Pete

2015-01-01

An examination is made of the hypothesis that the statistics of magnetic-storm-maximum intensities are the realization of a log-normal stochastic process. Weighted least-squares and maximum-likelihood methods are used to fit log-normal functions to −Dst storm-time maxima for years 1957-2012; bootstrap analysis is used to established confidence limits on forecasts. Both methods provide fits that are reasonably consistent with the data; both methods also provide fits that are superior to those that can be made with a power-law function. In general, the maximum-likelihood method provides forecasts having tighter confidence intervals than those provided by weighted least-squares. From extrapolation of maximum-likelihood fits: a magnetic storm with intensity exceeding that of the 1859 Carrington event, −Dst≥850 nT, occurs about 1.13 times per century and a wide 95% confidence interval of [0.42,2.41] times per century; a 100-yr magnetic storm is identified as having a −Dst≥880 nT (greater than Carrington) but a wide 95% confidence interval of [490,1187] nT.

A removal model for estimating detection probabilities from point-count surveys

USGS Publications Warehouse

Farnsworth, G.L.; Pollock, K.H.; Nichols, J.D.; Simons, T.R.; Hines, J.E.; Sauer, J.R.

2002-01-01

Use of point-count surveys is a popular method for collecting data on abundance and distribution of birds. However, analyses of such data often ignore potential differences in detection probability. We adapted a removal model to directly estimate detection probability during point-count surveys. The model assumes that singing frequency is a major factor influencing probability of detection when birds are surveyed using point counts. This may be appropriate for surveys in which most detections are by sound. The model requires counts to be divided into several time intervals. Point counts are often conducted for 10 min, where the number of birds recorded is divided into those first observed in the first 3 min, the subsequent 2 min, and the last 5 min. We developed a maximum-likelihood estimator for the detectability of birds recorded during counts divided into those intervals. This technique can easily be adapted to point counts divided into intervals of any length. We applied this method to unlimited-radius counts conducted in Great Smoky Mountains National Park. We used model selection criteria to identify whether detection probabilities varied among species, throughout the morning, throughout the season, and among different observers. We found differences in detection probability among species. Species that sing frequently such as Winter Wren (Troglodytes troglodytes) and Acadian Flycatcher (Empidonax virescens) had high detection probabilities (∼90%) and species that call infrequently such as Pileated Woodpecker (Dryocopus pileatus) had low detection probability (36%). We also found detection probabilities varied with the time of day for some species (e.g. thrushes) and between observers for other species. We used the same approach to estimate detection probability and density for a subset of the observations with limited-radius point counts.

Estimating parameter of Rayleigh distribution by using Maximum Likelihood method and Bayes method

NASA Astrophysics Data System (ADS)

Ardianti, Fitri; Sutarman

2018-01-01

In this paper, we use Maximum Likelihood estimation and Bayes method under some risk function to estimate parameter of Rayleigh distribution to know the best method. The prior knowledge which used in Bayes method is Jeffrey’s non-informative prior. Maximum likelihood estimation and Bayes method under precautionary loss function, entropy loss function, loss function-L 1 will be compared. We compare these methods by bias and MSE value using R program. After that, the result will be displayed in tables to facilitate the comparisons.

Consistency of Rasch Model Parameter Estimation: A Simulation Study.

ERIC Educational Resources Information Center

van den Wollenberg, Arnold L.; And Others

1988-01-01

The unconditional--simultaneous--maximum likelihood (UML) estimation procedure for the one-parameter logistic model produces biased estimators. The UML method is inconsistent and is not a good alternative to conditional maximum likelihood method, at least with small numbers of items. The minimum Chi-square estimation procedure produces unbiased…

Maximum Likelihood Estimations and EM Algorithms with Length-biased Data

PubMed Central

Qin, Jing; Ning, Jing; Liu, Hao; Shen, Yu

2012-01-01

SUMMARY Length-biased sampling has been well recognized in economics, industrial reliability, etiology applications, epidemiological, genetic and cancer screening studies. Length-biased right-censored data have a unique data structure different from traditional survival data. The nonparametric and semiparametric estimations and inference methods for traditional survival data are not directly applicable for length-biased right-censored data. We propose new expectation-maximization algorithms for estimations based on full likelihoods involving infinite dimensional parameters under three settings for length-biased data: estimating nonparametric distribution function, estimating nonparametric hazard function under an increasing failure rate constraint, and jointly estimating baseline hazards function and the covariate coefficients under the Cox proportional hazards model. Extensive empirical simulation studies show that the maximum likelihood estimators perform well with moderate sample sizes and lead to more efficient estimators compared to the estimating equation approaches. The proposed estimates are also more robust to various right-censoring mechanisms. We prove the strong consistency properties of the estimators, and establish the asymptotic normality of the semi-parametric maximum likelihood estimators under the Cox model using modern empirical processes theory. We apply the proposed methods to a prevalent cohort medical study. Supplemental materials are available online. PMID:22323840

Regression estimators for generic health-related quality of life and quality-adjusted life years.

PubMed

Basu, Anirban; Manca, Andrea

2012-01-01

To develop regression models for outcomes with truncated supports, such as health-related quality of life (HRQoL) data, and account for features typical of such data such as a skewed distribution, spikes at 1 or 0, and heteroskedasticity. Regression estimators based on features of the Beta distribution. First, both a single equation and a 2-part model are presented, along with estimation algorithms based on maximum-likelihood, quasi-likelihood, and Bayesian Markov-chain Monte Carlo methods. A novel Bayesian quasi-likelihood estimator is proposed. Second, a simulation exercise is presented to assess the performance of the proposed estimators against ordinary least squares (OLS) regression for a variety of HRQoL distributions that are encountered in practice. Finally, the performance of the proposed estimators is assessed by using them to quantify the treatment effect on QALYs in the EVALUATE hysterectomy trial. Overall model fit is studied using several goodness-of-fit tests such as Pearson's correlation test, link and reset tests, and a modified Hosmer-Lemeshow test. The simulation results indicate that the proposed methods are more robust in estimating covariate effects than OLS, especially when the effects are large or the HRQoL distribution has a large spike at 1. Quasi-likelihood techniques are more robust than maximum likelihood estimators. When applied to the EVALUATE trial, all but the maximum likelihood estimators produce unbiased estimates of the treatment effect. One and 2-part Beta regression models provide flexible approaches to regress the outcomes with truncated supports, such as HRQoL, on covariates, after accounting for many idiosyncratic features of the outcomes distribution. This work will provide applied researchers with a practical set of tools to model outcomes in cost-effectiveness analysis.

How Affiliation Disclosure and Control Over User-Generated Comments Affects Consumer Health Knowledge and Behavior: A Randomized Controlled Experiment of Pharmaceutical Direct-to-Consumer Advertising on Social Media.

PubMed

DeAndrea, David Christopher; Vendemia, Megan Ashley

2016-07-19

More people are seeking health information online than ever before and pharmaceutical companies are increasingly marketing their drugs through social media. The aim was to examine two major concerns related to online direct-to-consumer pharmaceutical advertising: (1) how disclosing an affiliation with a pharmaceutical company affects how people respond to drug information produced by both health organizations and online commenters, and (2) how knowledge that health organizations control the display of user-generated comments affects consumer health knowledge and behavior. We conducted a 2×2×2 between-subjects experiment (N=674). All participants viewed an infographic posted to Facebook by a health organization about a prescription allergy drug. Across conditions, the infographic varied in the degree to which the health organization and commenters appeared to be affiliated with a drug manufacturer, and the display of user-generated comments appeared to be controlled. Affiliation disclosure statements on a health organization's Facebook post increased perceptions of an organization-drug manufacturer connection, which reduced trust in the organization (point estimate -0.45, 95% CI -0.69 to -0.24) and other users who posted comments about the drug (point estimate -0.44, 95% CI -0.68 to -0.22). Furthermore, increased perceptions of an organization-manufacturer connection reduced the likelihood that people would recommend the drug to important others (point estimate -0.35, 95% CI -0.59 to -0.15), and share the drug post with others on Facebook (point estimate -0.37, 95% CI -0.64 to -0.16). An affiliation cue next to the commenters' names increased perceptions that the commenters were affiliated with the drug manufacturer, which reduced trust in the comments (point estimate -0.81, 95% CI -1.04 to -0.59), the organization that made the post (point estimate -0.68, 95% CI -0.90 to -0.49), the likelihood of participants recommending the drug (point estimate -0.61, 95% CI -0.82 to -0.43), and sharing the post with others on Facebook (point estimate -0.63, 95% CI -0.87 to -0.43). Cues indicating that a health organization removed user-generated comments from a post increased perceptions that the drug manufacturer influenced the display of the comments, which negatively affected trust in the comments (point estimate -0.35, 95% CI -0.53 to -0.20), the organization (point estimate -0.31, 95% CI -0.47 to -0.17), the likelihood of recommending the drug (point estimate -0.26, 95% CI -0.41 to -0.14), and the likelihood of sharing the post with others on Facebook (point estimate -0.28, 95% CI -0.45 to -0.15). (All estimates are unstandardized indirect effects and 95% bias-corrected bootstrap confidence intervals.) Concern over pharmaceutical companies hiding their affiliations and strategically controlling user-generated comments is well founded; these practices can greatly affect not only how viewers evaluate drug information online, but also how likely they are to propagate the information throughout their online and offline social networks.

Statistical inference for tumor growth inhibition T/C ratio.

PubMed

Wu, Jianrong

2010-09-01

The tumor growth inhibition T/C ratio is commonly used to quantify treatment effects in drug screening tumor xenograft experiments. The T/C ratio is converted to an antitumor activity rating using an arbitrary cutoff point and often without any formal statistical inference. Here, we applied a nonparametric bootstrap method and a small sample likelihood ratio statistic to make a statistical inference of the T/C ratio, including both hypothesis testing and a confidence interval estimate. Furthermore, sample size and power are also discussed for statistical design of tumor xenograft experiments. Tumor xenograft data from an actual experiment were analyzed to illustrate the application.

A Power-Law Growth and Decay Model with Autocorrelation for Posting Data to Social Networking Services

PubMed Central

Fujiyama, Toshifumi; Matsui, Chihiro; Takemura, Akimichi

2016-01-01

We propose a power-law growth and decay model for posting data to social networking services before and after social events. We model the time series structure of deviations from the power-law growth and decay with a conditional Poisson autoregressive (AR) model. Online postings related to social events are described by five parameters in the power-law growth and decay model, each of which characterizes different aspects of interest in the event. We assess the validity of parameter estimates in terms of confidence intervals, and compare various submodels based on likelihoods and information criteria. PMID:27505155

High-Performance Clock Synchronization Algorithms for Distributed Wireless Airborne Computer Networks with Applications to Localization and Tracking of Targets

DTIC Science & Technology

2010-06-01

GMKPF represents a better and more flexible alternative to the Gaussian Maximum Likelihood (GML), and Exponential Maximum Likelihood ( EML ...accurate results relative to GML and EML when the network delays are modeled in terms of a single non-Gaussian/non-exponential distribution or as a...to the Gaussian Maximum Likelihood (GML), and Exponential Maximum Likelihood ( EML ) estimators for clock offset estimation in non-Gaussian or non

Maximum-likelihood estimation of parameterized wavefronts from multifocal data

PubMed Central

Sakamoto, Julia A.; Barrett, Harrison H.

2012-01-01

A method for determining the pupil phase distribution of an optical system is demonstrated. Coefficients in a wavefront expansion were estimated using likelihood methods, where the data consisted of multiple irradiance patterns near focus. Proof-of-principle results were obtained in both simulation and experiment. Large-aberration wavefronts were handled in the numerical study. Experimentally, we discuss the handling of nuisance parameters. Fisher information matrices, Cramér-Rao bounds, and likelihood surfaces are examined. ML estimates were obtained by simulated annealing to deal with numerous local extrema in the likelihood function. Rapid processing techniques were employed to reduce the computational time. PMID:22772282

A comparison of abundance estimates from extended batch-marking and Jolly–Seber-type experiments

PubMed Central

Cowen, Laura L E; Besbeas, Panagiotis; Morgan, Byron J T; Schwarz, Carl J

2014-01-01

Little attention has been paid to the use of multi-sample batch-marking studies, as it is generally assumed that an individual's capture history is necessary for fully efficient estimates. However, recently, Huggins et al. (2010) present a pseudo-likelihood for a multi-sample batch-marking study where they used estimating equations to solve for survival and capture probabilities and then derived abundance estimates using a Horvitz–Thompson-type estimator. We have developed and maximized the likelihood for batch-marking studies. We use data simulated from a Jolly–Seber-type study and convert this to what would have been obtained from an extended batch-marking study. We compare our abundance estimates obtained from the Crosbie–Manly–Arnason–Schwarz (CMAS) model with those of the extended batch-marking model to determine the efficiency of collecting and analyzing batch-marking data. We found that estimates of abundance were similar for all three estimators: CMAS, Huggins, and our likelihood. Gains are made when using unique identifiers and employing the CMAS model in terms of precision; however, the likelihood typically had lower mean square error than the pseudo-likelihood method of Huggins et al. (2010). When faced with designing a batch-marking study, researchers can be confident in obtaining unbiased abundance estimators. Furthermore, they can design studies in order to reduce mean square error by manipulating capture probabilities and sample size. PMID:24558576

Estimation of the incubation period of influenza A (H1N1-2009) among imported cases: addressing censoring using outbreak data at the origin of importation.

PubMed

Nishiura, Hiroshi; Inaba, Hisashi

2011-03-07

Empirical estimates of the incubation period of influenza A (H1N1-2009) have been limited. We estimated the incubation period among confirmed imported cases who traveled to Japan from Hawaii during the early phase of the 2009 pandemic (n=72). We addressed censoring and employed an infection-age structured argument to explicitly model the daily frequency of illness onset after departure. We assumed uniform and exponential distributions for the frequency of exposure in Hawaii, and the hazard rate of infection for the latter assumption was retrieved, in Hawaii, from local outbreak data. The maximum likelihood estimates of the median incubation period range from 1.43 to 1.64 days according to different modeling assumptions, consistent with a published estimate based on a New York school outbreak. The likelihood values of the different modeling assumptions do not differ greatly from each other, although models with the exponential assumption yield slightly shorter incubation periods than those with the uniform exposure assumption. Differences between our proposed approach and a published method for doubly interval-censored analysis highlight the importance of accounting for the dependence of the frequency of exposure on the survival function of incubating individuals among imported cases. A truncation of the density function of the incubation period due to an absence of illness onset during the exposure period also needs to be considered. When the data generating process is similar to that among imported cases, and when the incubation period is close to or shorter than the length of exposure, accounting for these aspects is critical for long exposure times. Copyright © 2010 Elsevier Ltd. All rights reserved.

Profile-Likelihood Approach for Estimating Generalized Linear Mixed Models with Factor Structures

ERIC Educational Resources Information Center

Jeon, Minjeong; Rabe-Hesketh, Sophia

2012-01-01

In this article, the authors suggest a profile-likelihood approach for estimating complex models by maximum likelihood (ML) using standard software and minimal programming. The method works whenever setting some of the parameters of the model to known constants turns the model into a standard model. An important class of models that can be…

Multiple-Hit Parameter Estimation in Monolithic Detectors

PubMed Central

Barrett, Harrison H.; Lewellen, Tom K.; Miyaoka, Robert S.

2014-01-01

We examine a maximum-a-posteriori method for estimating the primary interaction position of gamma rays with multiple interaction sites (hits) in a monolithic detector. In assessing the performance of a multiple-hit estimator over that of a conventional one-hit estimator, we consider a few different detector and readout configurations of a 50-mm-wide square cerium-doped lutetium oxyorthosilicate block. For this study, we use simulated data from SCOUT, a Monte-Carlo tool for photon tracking and modeling scintillation- camera output. With this tool, we determine estimate bias and variance for a multiple-hit estimator and compare these with similar metrics for a one-hit maximum-likelihood estimator, which assumes full energy deposition in one hit. We also examine the effect of event filtering on these metrics; for this purpose, we use a likelihood threshold to reject signals that are not likely to have been produced under the assumed likelihood model. Depending on detector design, we observe a 1%–12% improvement of intrinsic resolution for a 1-or-2-hit estimator as compared with a 1-hit estimator. We also observe improved differentiation of photopeak events using a 1-or-2-hit estimator as compared with the 1-hit estimator; more than 6% of photopeak events that were rejected by likelihood filtering for the 1-hit estimator were accurately identified as photopeak events and positioned without loss of resolution by a 1-or-2-hit estimator; for PET, this equates to at least a 12% improvement in coincidence-detection efficiency with likelihood filtering applied. PMID:23193231

Supplement: “The Rate of Binary Black Hole Mergers Inferred from Advanced LIGO Observations Surrounding GW150914” (2016, ApJL, 833, L1)

NASA Astrophysics Data System (ADS)

Abbott, B. P.; Abbott, R.; Abbott, T. D.; Abernathy, M. R.; Acernese, F.; Ackley, K.; Adams, C.; Adams, T.; Addesso, P.; Adhikari, R. X.; Adya, V. B.; Affeldt, C.; Agathos, M.; Agatsuma, K.; Aggarwal, N.; Aguiar, O. D.; Aiello, L.; Ain, A.; Ajith, P.; Allen, B.; Allocca, A.; Altin, P. A.; Anderson, S. B.; Anderson, W. G.; Arai, K.; Araya, M. C.; Arceneaux, C. C.; Areeda, J. S.; Arnaud, N.; Arun, K. G.; Ascenzi, S.; Ashton, G.; Ast, M.; Aston, S. M.; Astone, P.; Aufmuth, P.; Aulbert, C.; Babak, S.; Bacon, P.; Bader, M. K. M.; Baker, P. T.; Baldaccini, F.; Ballardin, G.; Ballmer, S. W.; Barayoga, J. C.; Barclay, S. E.; Barish, B. C.; Barker, D.; Barone, F.; Barr, B.; Barsotti, L.; Barsuglia, M.; Barta, D.; Bartlett, J.; Bartos, I.; Bassiri, R.; Basti, A.; Batch, J. C.; Baune, C.; Bavigadda, V.; Bazzan, M.; Behnke, B.; Bejger, M.; Bell, A. S.; Bell, C. J.; Berger, B. K.; Bergman, J.; Bergmann, G.; Berry, C. P. L.; Bersanetti, D.; Bertolini, A.; Betzwieser, J.; Bhagwat, S.; Bhandare, R.; Bilenko, I. A.; Billingsley, G.; Birch, J.; Birney, R.; Biscans, S.; Bisht, A.; Bitossi, M.; Biwer, C.; Bizouard, M. A.; Blackburn, J. K.; Blair, C. D.; Blair, D. G.; Blair, R. M.; Bloemen, S.; Bock, O.; Bodiya, T. P.; Boer, M.; Bogaert, G.; Bogan, C.; Bohe, A.; Bojtos, P.; Bond, C.; Bondu, F.; Bonnand, R.; Boom, B. A.; Bork, R.; Boschi, V.; Bose, S.; Bouffanais, Y.; Bozzi, A.; Bradaschia, C.; Brady, P. R.; Braginsky, V. B.; Branchesi, M.; Brau, J. E.; Briant, T.; Brillet, A.; Brinkmann, M.; Brisson, V.; Brockill, P.; Brooks, A. F.; Brown, D. A.; Brown, D. D.; Brown, N. M.; Buchanan, C. C.; Buikema, A.; Bulik, T.; Bulten, H. J.; Buonanno, A.; Buskulic, D.; Buy, C.; Byer, R. L.; Cadonati, L.; Cagnoli, G.; Cahillane, C.; Calderón Bustillo, J.; Callister, T.; Calloni, E.; Camp, J. B.; Cannon, K. C.; Cao, J.; Capano, C. D.; Capocasa, E.; Carbognani, F.; Caride, S.; Casanueva Diaz, J.; Casentini, C.; Caudill, S.; Cavaglià, M.; Cavalier, F.; Cavalieri, R.; Cella, G.; Cepeda, C. B.; Cerboni Baiardi, L.; Cerretani, G.; Cesarini, E.; Chakraborty, R.; Chalermsongsak, T.; Chamberlin, S. J.; Chan, M.; Chao, S.; Charlton, P.; Chassande-Mottin, E.; Chen, H. Y.; Chen, Y.; Cheng, C.; Chincarini, A.; Chiummo, A.; Cho, H. S.; Cho, M.; Chow, J. H.; Christensen, N.; Chu, Q.; Chua, S.; Chung, S.; Ciani, G.; Clara, F.; Clark, J. A.; Cleva, F.; Coccia, E.; Cohadon, P.-F.; Colla, A.; Collette, C. G.; Cominsky, L.; Constancio, M., Jr.; Conte, A.; Conti, L.; Cook, D.; Corbitt, T. R.; Cornish, N.; Corsi, A.; Cortese, S.; Costa, C. A.; Coughlin, M. W.; Coughlin, S. B.; Coulon, J.-P.; Countryman, S. T.; Couvares, P.; Cowan, E. E.; Coward, D. M.; Cowart, M. J.; Coyne, D. C.; Coyne, R.; Craig, K.; Creighton, J. D. E.; Cripe, J.; Crowder, S. G.; Cumming, A.; Cunningham, L.; Cuoco, E.; Dal Canton, T.; Danilishin, S. L.; D’Antonio, S.; Danzmann, K.; Darman, N. S.; Dattilo, V.; Dave, I.; Daveloza, H. P.; Davier, M.; Davies, G. S.; Daw, E. J.; Day, R.; De, S.; DeBra, D.; Debreczeni, G.; Degallaix, J.; De Laurentis, M.; Deléglise, S.; Del Pozzo, W.; Denker, T.; Dent, T.; Dereli, H.; Dergachev, V.; De Rosa, R.; DeRosa, R. T.; DeSalvo, R.; Dhurandhar, S.; Díaz, M. C.; Di Fiore, L.; Di Giovanni, M.; Di Lieto, A.; Di Pace, S.; Di Palma, I.; Di Virgilio, A.; Dojcinoski, G.; Dolique, V.; Donovan, F.; Dooley, K. L.; Doravari, S.; Douglas, R.; Downes, T. P.; Drago, M.; Drever, R. W. P.; Driggers, J. C.; Du, Z.; Ducrot, M.; Dwyer, S. E.; Edo, T. B.; Edwards, M. C.; Effler, A.; Eggenstein, H.-B.; Ehrens, P.; Eichholz, J.; Eikenberry, S. S.; Engels, W.; Essick, R. C.; Etzel, T.; Evans, M.; Evans, T. M.; Everett, R.; Factourovich, M.; Fafone, V.; Fair, H.; Fairhurst, S.; Fan, X.; Fang, Q.; Farinon, S.; Farr, B.; Farr, W. M.; Favata, M.; Fays, M.; Fehrmann, H.; Fejer, M. M.; Ferrante, I.; Ferreira, E. C.; Ferrini, F.; Fidecaro, F.; Fiori, I.; Fiorucci, D.; Fisher, R. P.; Flaminio, R.; Fletcher, M.; Fong, H.; Fournier, J.-D.; Franco, S.; Frasca, S.; Frasconi, F.; Frei, Z.; Freise, A.; Frey, R.; Frey, V.; Fricke, T. T.; Fritschel, P.; Frolov, V. V.; Fulda, P.; Fyffe, M.; Gabbard, H. A. G.; Gair, J. R.; Gammaitoni, L.; Gaonkar, S. G.; Garufi, F.; Gatto, A.; Gaur, G.; Gehrels, N.; Gemme, G.; Gendre, B.; Genin, E.; Gennai, A.; George, J.; Gergely, L.; Germain, V.; Ghosh, Archisman; Ghosh, S.; Giaime, J. A.; Giardina, K. D.; Giazotto, A.; Gill, K.; Glaefke, A.; Goetz, E.; Goetz, R.; Gondan, L.; González, G.; Gonzalez Castro, J. M.; Gopakumar, A.; Gordon, N. A.; Gorodetsky, M. L.; Gossan, S. E.; Gosselin, M.; Gouaty, R.; Graef, C.; Graff, P. B.; Granata, M.; Grant, A.; Gras, S.; Gray, C.; Greco, G.; Green, A. C.; Groot, P.; Grote, H.; Grunewald, S.; Guidi, G. M.; Guo, X.; Gupta, A.; Gupta, M. K.; Gushwa, K. E.; Gustafson, E. K.; Gustafson, R.; Hacker, J. J.; Hall, B. R.; Hall, E. D.; Hammond, G.; Haney, M.; Hanke, M. M.; Hanks, J.; Hanna, C.; Hannam, M. D.; Hanson, J.; Hardwick, T.; Harms, J.; Harry, G. M.; Harry, I. W.; Hart, M. J.; Hartman, M. T.; Haster, C.-J.; Haughian, K.; Heidmann, A.; Heintze, M. C.; Heitmann, H.; Hello, P.; Hemming, G.; Hendry, M.; Heng, I. S.; Hennig, J.; Heptonstall, A. W.; Heurs, M.; Hild, S.; Hoak, D.; Hodge, K. A.; Hofman, D.; Hollitt, S. E.; Holt, K.; Holz, D. E.; Hopkins, P.; Hosken, D. J.; Hough, J.; Houston, E. A.; Howell, E. J.; Hu, Y. M.; Huang, S.; Huerta, E. A.; Huet, D.; Hughey, B.; Husa, S.; Huttner, S. H.; Huynh-Dinh, T.; Idrisy, A.; Indik, N.; Ingram, D. R.; Inta, R.; Isa, H. N.; Isac, J.-M.; Isi, M.; Islas, G.; Isogai, T.; Iyer, B. R.; Izumi, K.; Jacqmin, T.; Jang, H.; Jani, K.; Jaranowski, P.; Jawahar, S.; Jiménez-Forteza, F.; Johnson, W. W.; Jones, D. I.; Jones, R.; Jonker, R. J. G.; Ju, L.; K, Haris; Kalaghatgi, C. V.; Kalogera, V.; Kandhasamy, S.; Kang, G.; Kanner, J. B.; Karki, S.; Kasprzack, M.; Katsavounidis, E.; Katzman, W.; Kaufer, S.; Kaur, T.; Kawabe, K.; Kawazoe, F.; Kéfélian, F.; Kehl, M. S.; Keitel, D.; Kelley, D. B.; Kells, W.; Kennedy, R.; Key, J. S.; Khalaidovski, A.; Khalili, F. Y.; Khan, I.; Khan, S.; Khan, Z.; Khazanov, E. A.; Kijbunchoo, N.; Kim, C.; Kim, J.; Kim, K.; Kim, Nam-Gyu; Kim, Namjun; Kim, Y.-M.; King, E. J.; King, P. J.; Kinzel, D. L.; Kissel, J. S.; Kleybolte, L.; Klimenko, S.; Koehlenbeck, S. M.; Kokeyama, K.; Koley, S.; Kondrashov, V.; Kontos, A.; Korobko, M.; Korth, W. Z.; Kowalska, I.; Kozak, D. B.; Kringel, V.; Krishnan, B.; Królak, A.; Krueger, C.; Kuehn, G.; Kumar, P.; Kuo, L.; Kutynia, A.; Lackey, B. D.; Landry, M.; Lange, J.; Lantz, B.; Lasky, P. D.; Lazzarini, A.; Lazzaro, C.; Leaci, P.; Leavey, S.; Lebigot, E. O.; Lee, C. H.; Lee, H. K.; Lee, H. M.; Lee, K.; Lenon, A.; Leonardi, M.; Leong, J. R.; Leroy, N.; Letendre, N.; Levin, Y.; Levine, B. M.; Li, T. G. F.; Libson, A.; Littenberg, T. B.; Lockerbie, N. A.; Logue, J.; Lombardi, A. L.; Lord, J. E.; Lorenzini, M.; Loriette, V.; Lormand, M.; Losurdo, G.; Lough, J. D.; Lück, H.; Lundgren, A. P.; Luo, J.; Lynch, R.; Ma, Y.; MacDonald, T.; Machenschalk, B.; MacInnis, M.; Macleod, D. M.; Magaña-Sandoval, F.; Magee, R. M.; Mageswaran, M.; Majorana, E.; Maksimovic, I.; Malvezzi, V.; Man, N.; Mandel, I.; Mandic, V.; Mangano, V.; Mansell, G. L.; Manske, M.; Mantovani, M.; Marchesoni, F.; Marion, F.; Márka, S.; Márka, Z.; Markosyan, A. S.; Maros, E.; Martelli, F.; Martellini, L.; Martin, I. W.; Martin, R. M.; Martynov, D. V.; Marx, J. N.; Mason, K.; Masserot, A.; Massinger, T. J.; Masso-Reid, M.; Matichard, F.; Matone, L.; Mavalvala, N.; Mazumder, N.; Mazzolo, G.; McCarthy, R.; McClelland, D. E.; McCormick, S.; McGuire, S. C.; McIntyre, G.; McIver, J.; McManus, D. J.; McWilliams, S. T.; Meacher, D.; Meadors, G. D.; Meidam, J.; Melatos, A.; Mendell, G.; Mendoza-Gandara, D.; Mercer, R. A.; Merilh, E.; Merzougui, M.; Meshkov, S.; Messenger, C.; Messick, C.; Meyers, P. M.; Mezzani, F.; Miao, H.; Michel, C.; Middleton, H.; Mikhailov, E. E.; Milano, L.; Miller, J.; Millhouse, M.; Minenkov, Y.; Ming, J.; Mirshekari, S.; Mishra, C.; Mitra, S.; Mitrofanov, V. P.; Mitselmakher, G.; Mittleman, R.; Moggi, A.; Mohan, M.; Mohapatra, S. R. P.; Montani, M.; Moore, B. C.; Moore, C. J.; Moraru, D.; Moreno, G.; Morriss, S. R.; Mossavi, K.; Mours, B.; Mow-Lowry, C. M.; Mueller, C. L.; Mueller, G.; Muir, A. W.; Mukherjee, Arunava; Mukherjee, D.; Mukherjee, S.; Mukund, N.; Mullavey, A.; Munch, J.; Murphy, D. J.; Murray, P. G.; Mytidis, A.; Nardecchia, I.; Naticchioni, L.; Nayak, R. K.; Necula, V.; Nedkova, K.; Nelemans, G.; Neri, M.; Neunzert, A.; Newton, G.; Nguyen, T. T.; Nielsen, A. B.; Nissanke, S.; Nitz, A.; Nocera, F.; Nolting, D.; Normandin, M. E.; Nuttall, L. K.; Oberling, J.; Ochsner, E.; O’Dell, J.; Oelker, E.; Ogin, G. H.; Oh, J. J.; Oh, S. H.; Ohme, F.; Oliver, M.; Oppermann, P.; Oram, Richard J.; O’Reilly, B.; O’Shaughnessy, R.; Ottaway, D. J.; Ottens, R. S.; Overmier, H.; Owen, B. J.; Pai, A.; Pai, S. A.; Palamos, J. R.; Palashov, O.; Palomba, C.; Pal-Singh, A.; Pan, H.; Pankow, C.; Pannarale, F.; Pant, B. C.; Paoletti, F.; Paoli, A.; Papa, M. A.; Paris, H. R.; Parker, W.; Pascucci, D.; Pasqualetti, A.; Passaquieti, R.; Passuello, D.; Patricelli, B.; Patrick, Z.; Pearlstone, B. L.; Pedraza, M.; Pedurand, R.; Pekowsky, L.; Pele, A.; Penn, S.; Perreca, A.; Phelps, M.; Piccinni, O.; Pichot, M.; Piergiovanni, F.; Pierro, V.; Pillant, G.; Pinard, L.; Pinto, I. M.; Pitkin, M.; Poggiani, R.; Popolizio, P.; Porter, E. K.; Post, A.; Powell, J.; Prasad, J.; Predoi, V.; Premachandra, S. S.; Prestegard, T.; Price, L. R.; Prijatelj, M.; Principe, M.; Privitera, S.; Prodi, G. A.; Prokhorov, L.; Puncken, O.; Punturo, M.; Puppo, P.; Pürrer, M.; Qi, H.; Qin, J.; Quetschke, V.; Quintero, E. A.; Quitzow-James, R.; Raab, F. J.; Rabeling, D. S.; Radkins, H.; Raffai, P.; Raja, S.; Rakhmanov, M.; Rapagnani, P.; Raymond, V.; Razzano, M.; Re, V.; Read, J.; Reed, C. M.; Regimbau, T.; Rei, L.; Reid, S.; Reitze, D. H.; Rew, H.; Reyes, S. D.; Ricci, F.; Riles, K.; Robertson, N. A.; Robie, R.; Robinet, F.; Rocchi, A.; Rolland, L.; Rollins, J. G.; Roma, V. J.; Romano, R.; Romanov, G.; Romie, J. H.; Rosińska, D.; Rowan, S.; Rüdiger, A.; Ruggi, P.; Ryan, K.; Sachdev, S.; Sadecki, T.; Sadeghian, L.; Salconi, L.; Saleem, M.; Salemi, F.; Samajdar, A.; Sammut, L.; Sampson, L.; Sanchez, E. J.; Sandberg, V.; Sandeen, B.; Sanders, J. R.; Sassolas, B.; Sathyaprakash, B. S.; Saulson, P. R.; Sauter, O.; Savage, R. L.; Sawadsky, A.; Schale, P.; Schilling, R.; Schmidt, J.; Schmidt, P.; Schnabel, R.; Schofield, R. M. S.; Schönbeck, A.; Schreiber, E.; Schuette, D.; Schutz, B. F.; Scott, J.; Scott, S. M.; Sellers, D.; Sengupta, A. S.; Sentenac, D.; Sequino, V.; Sergeev, A.; Serna, G.; Setyawati, Y.; Sevigny, A.; Shaddock, D. A.; Shah, S.; Shahriar, M. S.; Shaltev, M.; Shao, Z.; Shapiro, B.; Shawhan, P.; Sheperd, A.; Shoemaker, D. H.; Shoemaker, D. M.; Siellez, K.; Siemens, X.; Sigg, D.; Silva, A. D.; Simakov, D.; Singer, A.; Singer, L. P.; Singh, A.; Singh, R.; Singhal, A.; Sintes, A. M.; Slagmolen, B. J. J.; Smith, J. R.; Smith, N. D.; Smith, R. J. E.; Son, E. J.; Sorazu, B.; Sorrentino, F.; Souradeep, T.; Srivastava, A. K.; Staley, A.; Steinke, M.; Steinlechner, J.; Steinlechner, S.; Steinmeyer, D.; Stephens, B. C.; Stevenson, S.; Stone, R.; Strain, K. A.; Straniero, N.; Stratta, G.; Strauss, N. A.; Strigin, S.; Sturani, R.; Stuver, A. L.; Summerscales, T. Z.; Sun, L.; Sutton, P. J.; Swinkels, B. L.; Szczepańczyk, M. J.; Tacca, M.; Talukder, D.; Tanner, D. B.; Tápai, M.; Tarabrin, S. P.; Taracchini, A.; Taylor, R.; Theeg, T.; Thirugnanasambandam, M. P.; Thomas, E. G.; Thomas, M.; Thomas, P.; Thorne, K. A.; Thorne, K. S.; Thrane, E.; Tiwari, S.; Tiwari, V.; Tokmakov, K. V.; Tomlinson, C.; Tonelli, M.; Torres, C. V.; Torrie, C. I.; Töyrä, D.; Travasso, F.; Traylor, G.; Trifirò, D.; Tringali, M. C.; Trozzo, L.; Tse, M.; Turconi, M.; Tuyenbayev, D.; Ugolini, D.; Unnikrishnan, C. S.; Urban, A. L.; Usman, S. A.; Vahlbruch, H.; Vajente, G.; Valdes, G.; Vallisneri, M.; van Bakel, N.; van Beuzekom, M.; van den Brand, J. F. J.; Van Den Broeck, C.; Vander-Hyde, D. C.; van der Schaaf, L.; van Heijningen, J. V.; van Veggel, A. A.; Vardaro, M.; Vass, S.; Vasúth, M.; Vaulin, R.; Vecchio, A.; Vedovato, G.; Veitch, J.; Veitch, P. J.; Venkateswara, K.; Verkindt, D.; Vetrano, F.; Viceré, A.; Vinciguerra, S.; Vine, D. J.; Vinet, J.-Y.; Vitale, S.; Vo, T.; Vocca, H.; Vorvick, C.; Voss, D.; Vousden, W. D.; Vyatchanin, S. P.; Wade, A. R.; Wade, L. E.; Wade, M.; Walker, M.; Wallace, L.; Walsh, S.; Wang, G.; Wang, H.; Wang, M.; Wang, X.; Wang, Y.; Ward, R. L.; Warner, J.; Was, M.; Weaver, B.; Wei, L.-W.; Weinert, M.; Weinstein, A. J.; Weiss, R.; Welborn, T.; Wen, L.; Wesels, P.; Westphal, T.; Wette, K.; Whelan, J. T.; White, D. J.; Whiting, B. F.; Williams, R. D.; Williamson, A. R.; Willis, J. L.; Willke, B.; Wimmer, M. H.; Winkler, W.; Wipf, C. C.; Wittel, H.; Woan, G.; Worden, J.; Wright, J. L.; Wu, G.; Yablon, J.; Yam, W.; Yamamoto, H.; Yancey, C. C.; Yap, M. J.; Yu, H.; Yvert, M.; Zadrożny, A.; Zangrando, L.; Zanolin, M.; Zendri, J.-P.; Zevin, M.; Zhang, F.; Zhang, L.; Zhang, M.; Zhang, Y.; Zhao, C.; Zhou, M.; Zhou, Z.; Zhu, X. J.; Zucker, M. E.; Zuraw, S. E.; Zweizig, J.; LIGO Scientific Collaboration; Virgo Collaboration

2016-12-01

This article provides supplemental information for a Letter reporting the rate of (BBH) coalescences inferred from 16 days of coincident Advanced LIGO observations surrounding the transient (GW) signal GW150914. In that work we reported various rate estimates whose 90% confidence intervals fell in the range 2–600 Gpc‑3 yr‑1. Here we give details on our method and computations, including information about our search pipelines, a derivation of our likelihood function for the analysis, a description of the astrophysical search trigger distribution expected from merging BBHs, details on our computational methods, a description of the effects and our model for calibration uncertainty, and an analytic method for estimating our detector sensitivity, which is calibrated to our measurements.

The beta Burr type X distribution properties with application.

PubMed

Merovci, Faton; Khaleel, Mundher Abdullah; Ibrahim, Noor Akma; Shitan, Mahendran

2016-01-01

We develop a new continuous distribution called the beta-Burr type X distribution that extends the Burr type X distribution. The properties provide a comprehensive mathematical treatment of this distribution. Further more, various structural properties of the new distribution are derived, that includes moment generating function and the rth moment thus generalizing some results in the literature. We also obtain expressions for the density, moment generating function and rth moment of the order statistics. We consider the maximum likelihood estimation to estimate the parameters. Additionally, the asymptotic confidence intervals for the parameters are derived from the Fisher information matrix. Finally, simulation study is carried at under varying sample size to assess the performance of this model. Illustration the real dataset indicates that this new distribution can serve as a good alternative model to model positive real data in many areas.

Unified framework to evaluate panmixia and migration direction among multiple sampling locations.

PubMed

Beerli, Peter; Palczewski, Michal

2010-05-01

For many biological investigations, groups of individuals are genetically sampled from several geographic locations. These sampling locations often do not reflect the genetic population structure. We describe a framework using marginal likelihoods to compare and order structured population models, such as testing whether the sampling locations belong to the same randomly mating population or comparing unidirectional and multidirectional gene flow models. In the context of inferences employing Markov chain Monte Carlo methods, the accuracy of the marginal likelihoods depends heavily on the approximation method used to calculate the marginal likelihood. Two methods, modified thermodynamic integration and a stabilized harmonic mean estimator, are compared. With finite Markov chain Monte Carlo run lengths, the harmonic mean estimator may not be consistent. Thermodynamic integration, in contrast, delivers considerably better estimates of the marginal likelihood. The choice of prior distributions does not influence the order and choice of the better models when the marginal likelihood is estimated using thermodynamic integration, whereas with the harmonic mean estimator the influence of the prior is pronounced and the order of the models changes. The approximation of marginal likelihood using thermodynamic integration in MIGRATE allows the evaluation of complex population genetic models, not only of whether sampling locations belong to a single panmictic population, but also of competing complex structured population models.

Analyzing animal movements using Brownian bridges.

PubMed

Horne, Jon S; Garton, Edward O; Krone, Stephen M; Lewis, Jesse S

2007-09-01

By studying animal movements, researchers can gain insight into many of the ecological characteristics and processes important for understanding population-level dynamics. We developed a Brownian bridge movement model (BBMM) for estimating the expected movement path of an animal, using discrete location data obtained at relatively short time intervals. The BBMM is based on the properties of a conditional random walk between successive pairs of locations, dependent on the time between locations, the distance between locations, and the Brownian motion variance that is related to the animal's mobility. We describe two critical developments that enable widespread use of the BBMM, including a derivation of the model when location data are measured with error and a maximum likelihood approach for estimating the Brownian motion variance. After the BBMM is fitted to location data, an estimate of the animal's probability of occurrence can be generated for an area during the time of observation. To illustrate potential applications, we provide three examples: estimating animal home ranges, estimating animal migration routes, and evaluating the influence of fine-scale resource selection on animal movement patterns.

Dietary acid, age, and serum bicarbonate levels among adults in the United States.

PubMed

Amodu, Afolarin; Abramowitz, Matthew K

2013-12-01

Greater dietary acid has been associated with lower serum bicarbonate levels in patients with CKD. Whether this association extends to the general population and if it is modified by age are unknown. This study examined the association of the dietary acid load, estimated by net endogenous acid production, with serum bicarbonate levels in adult participants in the National Health and Nutrition Examination Survey 1999-2004. The mean serum bicarbonate was 24.9 mEq/L (SEM=0.1), and the mean estimated net endogenous acid production was 57.4 mEq/d (SEM=0.4). Serum bicarbonate was linearly associated with age, such that the oldest participants had the highest serum bicarbonate levels. After multivariable adjustment, participants in the highest quartile of net endogenous acid production had 0.40 mEq/L (95% confidence interval, -0.55 to -0.26) lower serum bicarbonate and a 33% (95% confidence interval, 3 to 72) higher likelihood of acidosis compared with those participants in the lowest quartile. There was a significant interaction by age of the association of net endogenous acid production with serum bicarbonate (P=0.005). Among participants 20-39, 40-59, and ≥60 years old, those participants in the highest net endogenous acid production quartile had 0.26 (95% confidence interval, -0.49 to -0.03), 0.60 (95% confidence interval, -0.92 to -0.29), and 0.49 (95% confidence interval, -0.84 to -0.14) mEq/L lower serum bicarbonate, respectively, compared with participants in the lowest quartile. Greater dietary acid is associated with lower serum bicarbonate in the general US population, and the magnitude of this association is greater among middle-aged and elderly persons than younger adults.

Expected versus Observed Information in SEM with Incomplete Normal and Nonnormal Data

ERIC Educational Resources Information Center

Savalei, Victoria

2010-01-01

Maximum likelihood is the most common estimation method in structural equation modeling. Standard errors for maximum likelihood estimates are obtained from the associated information matrix, which can be estimated from the sample using either expected or observed information. It is known that, with complete data, estimates based on observed or…

An iterative procedure for obtaining maximum-likelihood estimates of the parameters for a mixture of normal distributions

NASA Technical Reports Server (NTRS)

Peters, B. C., Jr.; Walker, H. F.

1978-01-01

This paper addresses the problem of obtaining numerically maximum-likelihood estimates of the parameters for a mixture of normal distributions. In recent literature, a certain successive-approximations procedure, based on the likelihood equations, was shown empirically to be effective in numerically approximating such maximum-likelihood estimates; however, the reliability of this procedure was not established theoretically. Here, we introduce a general iterative procedure, of the generalized steepest-ascent (deflected-gradient) type, which is just the procedure known in the literature when the step-size is taken to be 1. We show that, with probability 1 as the sample size grows large, this procedure converges locally to the strongly consistent maximum-likelihood estimate whenever the step-size lies between 0 and 2. We also show that the step-size which yields optimal local convergence rates for large samples is determined in a sense by the 'separation' of the component normal densities and is bounded below by a number between 1 and 2.

An iterative procedure for obtaining maximum-likelihood estimates of the parameters for a mixture of normal distributions, 2

NASA Technical Reports Server (NTRS)

Peters, B. C., Jr.; Walker, H. F.

1976-01-01

The problem of obtaining numerically maximum likelihood estimates of the parameters for a mixture of normal distributions is addressed. In recent literature, a certain successive approximations procedure, based on the likelihood equations, is shown empirically to be effective in numerically approximating such maximum-likelihood estimates; however, the reliability of this procedure was not established theoretically. Here, a general iterative procedure is introduced, of the generalized steepest-ascent (deflected-gradient) type, which is just the procedure known in the literature when the step-size is taken to be 1. With probability 1 as the sample size grows large, it is shown that this procedure converges locally to the strongly consistent maximum-likelihood estimate whenever the step-size lies between 0 and 2. The step-size which yields optimal local convergence rates for large samples is determined in a sense by the separation of the component normal densities and is bounded below by a number between 1 and 2.

Bootstrap Standard Errors for Maximum Likelihood Ability Estimates When Item Parameters Are Unknown

ERIC Educational Resources Information Center

Patton, Jeffrey M.; Cheng, Ying; Yuan, Ke-Hai; Diao, Qi

2014-01-01

When item parameter estimates are used to estimate the ability parameter in item response models, the standard error (SE) of the ability estimate must be corrected to reflect the error carried over from item calibration. For maximum likelihood (ML) ability estimates, a corrected asymptotic SE is available, but it requires a long test and the…

Mortality table construction

NASA Astrophysics Data System (ADS)

Sutawanir

2015-12-01

Mortality tables play important role in actuarial studies such as life annuities, premium determination, premium reserve, valuation pension plan, pension funding. Some known mortality tables are CSO mortality table, Indonesian Mortality Table, Bowers mortality table, Japan Mortality table. For actuary applications some tables are constructed with different environment such as single decrement, double decrement, and multiple decrement. There exist two approaches in mortality table construction : mathematics approach and statistical approach. Distribution model and estimation theory are the statistical concepts that are used in mortality table construction. This article aims to discuss the statistical approach in mortality table construction. The distributional assumptions are uniform death distribution (UDD) and constant force (exponential). Moment estimation and maximum likelihood are used to estimate the mortality parameter. Moment estimation methods are easier to manipulate compared to maximum likelihood estimation (mle). However, the complete mortality data are not used in moment estimation method. Maximum likelihood exploited all available information in mortality estimation. Some mle equations are complicated and solved using numerical methods. The article focus on single decrement estimation using moment and maximum likelihood estimation. Some extension to double decrement will introduced. Simple dataset will be used to illustrated the mortality estimation, and mortality table.

Improving and Evaluating Nested Sampling Algorithm for Marginal Likelihood Estimation

NASA Astrophysics Data System (ADS)

Ye, M.; Zeng, X.; Wu, J.; Wang, D.; Liu, J.

2016-12-01

With the growing impacts of climate change and human activities on the cycle of water resources, an increasing number of researches focus on the quantification of modeling uncertainty. Bayesian model averaging (BMA) provides a popular framework for quantifying conceptual model and parameter uncertainty. The ensemble prediction is generated by combining each plausible model's prediction, and each model is attached with a model weight which is determined by model's prior weight and marginal likelihood. Thus, the estimation of model's marginal likelihood is crucial for reliable and accurate BMA prediction. Nested sampling estimator (NSE) is a new proposed method for marginal likelihood estimation. The process of NSE is accomplished by searching the parameters' space from low likelihood area to high likelihood area gradually, and this evolution is finished iteratively via local sampling procedure. Thus, the efficiency of NSE is dominated by the strength of local sampling procedure. Currently, Metropolis-Hasting (M-H) algorithm is often used for local sampling. However, M-H is not an efficient sampling algorithm for high-dimensional or complicated parameter space. For improving the efficiency of NSE, it could be ideal to incorporate the robust and efficient sampling algorithm - DREAMzs into the local sampling of NSE. The comparison results demonstrated that the improved NSE could improve the efficiency of marginal likelihood estimation significantly. However, both improved and original NSEs suffer from heavy instability. In addition, the heavy computation cost of huge number of model executions is overcome by using an adaptive sparse grid surrogates.

The Maximum Likelihood Estimation of Signature Transformation /MLEST/ algorithm. [for affine transformation of crop inventory data

NASA Technical Reports Server (NTRS)

Thadani, S. G.

1977-01-01

The Maximum Likelihood Estimation of Signature Transformation (MLEST) algorithm is used to obtain maximum likelihood estimates (MLE) of affine transformation. The algorithm has been evaluated for three sets of data: simulated (training and recognition segment pairs), consecutive-day (data gathered from Landsat images), and geographical-extension (large-area crop inventory experiment) data sets. For each set, MLEST signature extension runs were made to determine MLE values and the affine-transformed training segment signatures were used to classify the recognition segments. The classification results were used to estimate wheat proportions at 0 and 1% threshold values.

A threshold method for immunological correlates of protection

PubMed Central

2013-01-01

Background Immunological correlates of protection are biological markers such as disease-specific antibodies which correlate with protection against disease and which are measurable with immunological assays. It is common in vaccine research and in setting immunization policy to rely on threshold values for the correlate where the accepted threshold differentiates between individuals who are considered to be protected against disease and those who are susceptible. Examples where thresholds are used include development of a new generation 13-valent pneumococcal conjugate vaccine which was required in clinical trials to meet accepted thresholds for the older 7-valent vaccine, and public health decision making on vaccination policy based on long-term maintenance of protective thresholds for Hepatitis A, rubella, measles, Japanese encephalitis and others. Despite widespread use of such thresholds in vaccine policy and research, few statistical approaches have been formally developed which specifically incorporate a threshold parameter in order to estimate the value of the protective threshold from data. Methods We propose a 3-parameter statistical model called the a:b model which incorporates parameters for a threshold and constant but different infection probabilities below and above the threshold estimated using profile likelihood or least squares methods. Evaluation of the estimated threshold can be performed by a significance test for the existence of a threshold using a modified likelihood ratio test which follows a chi-squared distribution with 3 degrees of freedom, and confidence intervals for the threshold can be obtained by bootstrapping. The model also permits assessment of relative risk of infection in patients achieving the threshold or not. Goodness-of-fit of the a:b model may be assessed using the Hosmer-Lemeshow approach. The model is applied to 15 datasets from published clinical trials on pertussis, respiratory syncytial virus and varicella. Results Highly significant thresholds with p-values less than 0.01 were found for 13 of the 15 datasets. Considerable variability was seen in the widths of confidence intervals. Relative risks indicated around 70% or better protection in 11 datasets and relevance of the estimated threshold to imply strong protection. Goodness-of-fit was generally acceptable. Conclusions The a:b model offers a formal statistical method of estimation of thresholds differentiating susceptible from protected individuals which has previously depended on putative statements based on visual inspection of data. PMID:23448322

Predicted patient demand for a new delivery system for glaucoma medicine.

PubMed

Ozdemir, Semra; Wong, Tina T; Allingham, Robert Rand; Finkelstein, Eric A

2017-04-01

Sustained-release drug delivery systems that replace the need for daily glaucoma medications will improve outcomes for those who are nonadherent and reduce the inconvenience of having to take medications on a recurring basis.The objective is to estimate uptake (i.e., demand) for a new technology that delivers sustained-release glaucoma medication and to investigate how uptake varies by product attributes, physician recommendations, peer adoption (i.e., percentage of patients seen in a clinic using the new technology), and patient characteristics.In a web-enabled discrete-choice experiment survey, glaucoma patients in the United States were asked to choose between continuing eye drop use or purchasing the new delivery system. In a cross-sectional web-enabled survey, ophthalmologists were asked their likelihood of recommending the new technology based on product and patient characteristics.Study participants were 500 glaucoma patients who were on topical administration of daily eye drops and 155 ophthalmologists who practice in the US.Main outcomes were predicted uptake for patients and likelihood of recommending a new drug delivery system for ophthalmologists. Logistic models were used to analyze the choice data.Uptake was estimated to be 18% at an annual cost of $1000 and to be 24% when the cost was $500. A physician's recommendation increased uptake by 6% to 12%, whereas an increase in peer adoption from 5% to 50% increased uptake by 3% to 7%. Patients aged ≥ 65 and those with lower income were more likely to remain on eye drops. Physicians were more likely to recommend a product if the interval between administrations is 6 months or longer and when long-term safety and efficacy data are available. They were less likely to recommend it to patients with lower income and no adherence problems.Results suggest a significant interest in an injectable solution or other sustained-release alternatives to daily eye drops. However, in this survey, patient uptake was greatly influenced by out-of-pocket cost and the interval between treatment administrations. Few physicians were willing to recommend sustained-release technology if the treatment interval was less than 3 months.

A Statistical Method for Synthesizing Mediation Analyses Using the Product of Coefficient Approach Across Multiple Trials

PubMed Central

Huang, Shi; MacKinnon, David P.; Perrino, Tatiana; Gallo, Carlos; Cruden, Gracelyn; Brown, C Hendricks

2016-01-01

Mediation analysis often requires larger sample sizes than main effect analysis to achieve the same statistical power. Combining results across similar trials may be the only practical option for increasing statistical power for mediation analysis in some situations. In this paper, we propose a method to estimate: 1) marginal means for mediation path a, the relation of the independent variable to the mediator; 2) marginal means for path b, the relation of the mediator to the outcome, across multiple trials; and 3) the between-trial level variance-covariance matrix based on a bivariate normal distribution. We present the statistical theory and an R computer program to combine regression coefficients from multiple trials to estimate a combined mediated effect and confidence interval under a random effects model. Values of coefficients a and b, along with their standard errors from each trial are the input for the method. This marginal likelihood based approach with Monte Carlo confidence intervals provides more accurate inference than the standard meta-analytic approach. We discuss computational issues, apply the method to two real-data examples and make recommendations for the use of the method in different settings. PMID:28239330

Estimating stage-specific daily survival probabilities of nests when nest age is unknown

USGS Publications Warehouse

Stanley, T.R.

2004-01-01

Estimation of daily survival probabilities of nests is common in studies of avian populations. Since the introduction of Mayfield's (1961, 1975) estimator, numerous models have been developed to relax Mayfield's assumptions and account for biologically important sources of variation. Stanley (2000) presented a model for estimating stage-specific (e.g. incubation stage, nestling stage) daily survival probabilities of nests that conditions on “nest type” and requires that nests be aged when they are found. Because aging nests typically requires handling the eggs, there may be situations where nests can not or should not be aged and the Stanley (2000) model will be inapplicable. Here, I present a model for estimating stage-specific daily survival probabilities that conditions on nest stage for active nests, thereby obviating the need to age nests when they are found. Specifically, I derive the maximum likelihood function for the model, evaluate the model's performance using Monte Carlo simulations, and provide software for estimating parameters (along with an example). For sample sizes as low as 50 nests, bias was small and confidence interval coverage was close to the nominal rate, especially when a reduced-parameter model was used for estimation.

Asymptotic Properties of Induced Maximum Likelihood Estimates of Nonlinear Models for Item Response Variables: The Finite-Generic-Item-Pool Case.

ERIC Educational Resources Information Center

Jones, Douglas H.

The progress of modern mental test theory depends very much on the techniques of maximum likelihood estimation, and many popular applications make use of likelihoods induced by logistic item response models. While, in reality, item responses are nonreplicate within a single examinee and the logistic models are only ideal, practitioners make…

Computation of nonparametric convex hazard estimators via profile methods.

PubMed

Jankowski, Hanna K; Wellner, Jon A

2009-05-01

This paper proposes a profile likelihood algorithm to compute the nonparametric maximum likelihood estimator of a convex hazard function. The maximisation is performed in two steps: First the support reduction algorithm is used to maximise the likelihood over all hazard functions with a given point of minimum (or antimode). Then it is shown that the profile (or partially maximised) likelihood is quasi-concave as a function of the antimode, so that a bisection algorithm can be applied to find the maximum of the profile likelihood, and hence also the global maximum. The new algorithm is illustrated using both artificial and real data, including lifetime data for Canadian males and females.

Age-dependent biochemical quantities: an approach for calculating reference intervals.

PubMed

Bjerner, J

2007-01-01

A parametric method is often preferred when calculating reference intervals for biochemical quantities, as non-parametric methods are less efficient and require more observations/study subjects. Parametric methods are complicated, however, because of three commonly encountered features. First, biochemical quantities seldom display a Gaussian distribution, and there must either be a transformation procedure to obtain such a distribution or a more complex distribution has to be used. Second, biochemical quantities are often dependent on a continuous covariate, exemplified by rising serum concentrations of MUC1 (episialin, CA15.3) with increasing age. Third, outliers often exert substantial influence on parametric estimations and therefore need to be excluded before calculations are made. The International Federation of Clinical Chemistry (IFCC) currently recommends that confidence intervals be calculated for the reference centiles obtained. However, common statistical packages allowing for the adjustment of a continuous covariate do not make this calculation. In the method described in the current study, Tukey's fence is used to eliminate outliers and two-stage transformations (modulus-exponential-normal) in order to render Gaussian distributions. Fractional polynomials are employed to model functions for mean and standard deviations dependent on a covariate, and the model is selected by maximum likelihood. Confidence intervals are calculated for the fitted centiles by combining parameter estimation and sampling uncertainties. Finally, the elimination of outliers was made dependent on covariates by reiteration. Though a good knowledge of statistical theory is needed when performing the analysis, the current method is rewarding because the results are of practical use in patient care.

Applying a Weighted Maximum Likelihood Latent Trait Estimator to the Generalized Partial Credit Model

ERIC Educational Resources Information Center

Penfield, Randall D.; Bergeron, Jennifer M.

2005-01-01

This article applies a weighted maximum likelihood (WML) latent trait estimator to the generalized partial credit model (GPCM). The relevant equations required to obtain the WML estimator using the Newton-Raphson algorithm are presented, and a simulation study is described that compared the properties of the WML estimator to those of the maximum…

Rate of convergence of k-step Newton estimators to efficient likelihood estimators

Treesearch

Steve Verrill

2007-01-01

We make use of Cramer conditions together with the well-known local quadratic convergence of Newton?s method to establish the asymptotic closeness of k-step Newton estimators to efficient likelihood estimators. In Verrill and Johnson [2007. Confidence bounds and hypothesis tests for normal distribution coefficients of variation. USDA Forest Products Laboratory Research...

Collinear Latent Variables in Multilevel Confirmatory Factor Analysis: A Comparison of Maximum Likelihood and Bayesian Estimations.

PubMed

Can, Seda; van de Schoot, Rens; Hox, Joop

2015-06-01

Because variables may be correlated in the social and behavioral sciences, multicollinearity might be problematic. This study investigates the effect of collinearity manipulated in within and between levels of a two-level confirmatory factor analysis by Monte Carlo simulation. Furthermore, the influence of the size of the intraclass correlation coefficient (ICC) and estimation method; maximum likelihood estimation with robust chi-squares and standard errors and Bayesian estimation, on the convergence rate are investigated. The other variables of interest were rate of inadmissible solutions and the relative parameter and standard error bias on the between level. The results showed that inadmissible solutions were obtained when there was between level collinearity and the estimation method was maximum likelihood. In the within level multicollinearity condition, all of the solutions were admissible but the bias values were higher compared with the between level collinearity condition. Bayesian estimation appeared to be robust in obtaining admissible parameters but the relative bias was higher than for maximum likelihood estimation. Finally, as expected, high ICC produced less biased results compared to medium ICC conditions.

Improving regression-model-based streamwater constituent load estimates derived from serially correlated data

USGS Publications Warehouse

Aulenbach, Brent T.

2013-01-01

A regression-model based approach is a commonly used, efficient method for estimating streamwater constituent load when there is a relationship between streamwater constituent concentration and continuous variables such as streamwater discharge, season and time. A subsetting experiment using a 30-year dataset of daily suspended sediment observations from the Mississippi River at Thebes, Illinois, was performed to determine optimal sampling frequency, model calibration period length, and regression model methodology, as well as to determine the effect of serial correlation of model residuals on load estimate precision. Two regression-based methods were used to estimate streamwater loads, the Adjusted Maximum Likelihood Estimator (AMLE), and the composite method, a hybrid load estimation approach. While both methods accurately and precisely estimated loads at the model’s calibration period time scale, precisions were progressively worse at shorter reporting periods, from annually to monthly. Serial correlation in model residuals resulted in observed AMLE precision to be significantly worse than the model calculated standard errors of prediction. The composite method effectively improved upon AMLE loads for shorter reporting periods, but required a sampling interval of at least 15-days or shorter, when the serial correlations in the observed load residuals were greater than 0.15. AMLE precision was better at shorter sampling intervals and when using the shortest model calibration periods, such that the regression models better fit the temporal changes in the concentration–discharge relationship. The models with the largest errors typically had poor high flow sampling coverage resulting in unrepresentative models. Increasing sampling frequency and/or targeted high flow sampling are more efficient approaches to ensure sufficient sampling and to avoid poorly performing models, than increasing calibration period length.

IRT Item Parameter Recovery with Marginal Maximum Likelihood Estimation Using Loglinear Smoothing Models

ERIC Educational Resources Information Center

Casabianca, Jodi M.; Lewis, Charles

2015-01-01

Loglinear smoothing (LLS) estimates the latent trait distribution while making fewer assumptions about its form and maintaining parsimony, thus leading to more precise item response theory (IRT) item parameter estimates than standard marginal maximum likelihood (MML). This article provides the expectation-maximization algorithm for MML estimation…

An EM Algorithm for Maximum Likelihood Estimation of Process Factor Analysis Models

ERIC Educational Resources Information Center

Lee, Taehun

2010-01-01

In this dissertation, an Expectation-Maximization (EM) algorithm is developed and implemented to obtain maximum likelihood estimates of the parameters and the associated standard error estimates characterizing temporal flows for the latent variable time series following stationary vector ARMA processes, as well as the parameters defining the…

An Improved Nested Sampling Algorithm for Model Selection and Assessment

NASA Astrophysics Data System (ADS)

Zeng, X.; Ye, M.; Wu, J.; WANG, D.

2017-12-01

Multimodel strategy is a general approach for treating model structure uncertainty in recent researches. The unknown groundwater system is represented by several plausible conceptual models. Each alternative conceptual model is attached with a weight which represents the possibility of this model. In Bayesian framework, the posterior model weight is computed as the product of model prior weight and marginal likelihood (or termed as model evidence). As a result, estimating marginal likelihoods is crucial for reliable model selection and assessment in multimodel analysis. Nested sampling estimator (NSE) is a new proposed algorithm for marginal likelihood estimation. The implementation of NSE comprises searching the parameters' space from low likelihood area to high likelihood area gradually, and this evolution is finished iteratively via local sampling procedure. Thus, the efficiency of NSE is dominated by the strength of local sampling procedure. Currently, Metropolis-Hasting (M-H) algorithm and its variants are often used for local sampling in NSE. However, M-H is not an efficient sampling algorithm for high-dimensional or complex likelihood function. For improving the performance of NSE, it could be feasible to integrate more efficient and elaborated sampling algorithm - DREAMzs into the local sampling. In addition, in order to overcome the computation burden problem of large quantity of repeating model executions in marginal likelihood estimation, an adaptive sparse grid stochastic collocation method is used to build the surrogates for original groundwater model.

Hospitalization event notifications and reductions in readmissions of Medicare fee-for-service beneficiaries in the Bronx, New York.

PubMed

Unruh, Mark Aaron; Jung, Hye-Young; Kaushal, Rainu; Vest, Joshua R

2017-04-01

Follow-up with a primary care provider after hospital discharge has been associated with a reduced likelihood of readmission. However, primary care providers are frequently unaware of their patients' hospitalizations. Event notification may be an effective tool for reducing readmissions by notifying primary care providers when their patients have been admitted to and discharged from a hospital. We examined the effect of an event notification system on 30-day readmissions in the Bronx, New York. The Bronx has among the highest readmission rates in the country and is a particularly challenging setting to improve care due to the low socioeconomic status of the county and high rates of poor health behaviors among its residents. The study cohort included 2559 Medicare fee-for-service beneficiaries associated with 14 141 hospital admissions over the period January 2010 through June 2014. Linear regression models with beneficiary-level fixed-effects were used to estimate the impact of event notifications on readmissions by comparing the likelihood of rehospitalization for a beneficiary before and after event notifications were active. The unadjusted 30-day readmission rate when event notifications were not active was 29.5% compared to 26.5% when alerts were active. Regression estimates indicated that active hospitalization alert services were associated with a 2.9 percentage point reduction in the likelihood of readmission (95% confidence interval: -5.5, -0.4). Alerting providers through event notifications may be an effective tool for improving the quality and efficiency of care among high-risk populations. © The Author 2016. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Bayesian inference for dynamic transcriptional regulation; the Hes1 system as a case study.

PubMed

Heron, Elizabeth A; Finkenstädt, Bärbel; Rand, David A

2007-10-01

In this study, we address the problem of estimating the parameters of regulatory networks and provide the first application of Markov chain Monte Carlo (MCMC) methods to experimental data. As a case study, we consider a stochastic model of the Hes1 system expressed in terms of stochastic differential equations (SDEs) to which rigorous likelihood methods of inference can be applied. When fitting continuous-time stochastic models to discretely observed time series the lengths of the sampling intervals are important, and much of our study addresses the problem when the data are sparse. We estimate the parameters of an autoregulatory network providing results both for simulated and real experimental data from the Hes1 system. We develop an estimation algorithm using MCMC techniques which are flexible enough to allow for the imputation of latent data on a finer time scale and the presence of prior information about parameters which may be informed from other experiments as well as additional measurement error.

Multiple robustness in factorized likelihood models.

PubMed

Molina, J; Rotnitzky, A; Sued, M; Robins, J M

2017-09-01

We consider inference under a nonparametric or semiparametric model with likelihood that factorizes as the product of two or more variation-independent factors. We are interested in a finite-dimensional parameter that depends on only one of the likelihood factors and whose estimation requires the auxiliary estimation of one or several nuisance functions. We investigate general structures conducive to the construction of so-called multiply robust estimating functions, whose computation requires postulating several dimension-reducing models but which have mean zero at the true parameter value provided one of these models is correct.

Recovery of Item Parameters in the Nominal Response Model: A Comparison of Marginal Maximum Likelihood Estimation and Markov Chain Monte Carlo Estimation.

ERIC Educational Resources Information Center

Wollack, James A.; Bolt, Daniel M.; Cohen, Allan S.; Lee, Young-Sun

2002-01-01

Compared the quality of item parameter estimates for marginal maximum likelihood (MML) and Markov Chain Monte Carlo (MCMC) with the nominal response model using simulation. The quality of item parameter recovery was nearly identical for MML and MCMC, and both methods tended to produce good estimates. (SLD)

Vascular Disease, ESRD, and Death: Interpreting Competing Risk Analyses

PubMed Central

Coresh, Josef; Segev, Dorry L.; Kucirka, Lauren M.; Tighiouart, Hocine; Sarnak, Mark J.

2012-01-01

Summary Background and objectives Vascular disease, a common condition in CKD, is a risk factor for mortality and ESRD. Optimal patient care requires accurate estimation and ordering of these competing risks. Design, setting, participants, & measurements This is a prospective cohort study of screened (n=885) and randomized participants (n=837) in the Modification of Diet in Renal Disease study (original study enrollment, 1989–1992), evaluating the association of vascular disease with ESRD and pre-ESRD mortality using standard survival analysis and competing risk regression. Results The method of analysis resulted in markedly different estimates. Cumulative incidence by standard analysis (censoring at the competing event) implied that, with vascular disease, the 15-year incidence was 66% and 51% for ESRD and pre-ESRD death, respectively. A more accurate representation of absolute risk was estimated with competing risk regression: 15-year incidence was 54% and 29% for ESRD and pre-ESRD death, respectively. For the association of vascular disease with pre-ESRD death, estimates of relative risk by the two methods were similar (standard survival analysis adjusted hazard ratio, 1.63; 95% confidence interval, 1.20–2.20; competing risk regression adjusted subhazard ratio, 1.57; 95% confidence interval, 1.15–2.14). In contrast, the hazard and subhazard ratios differed substantially for other associations, such as GFR and pre-ESRD mortality. Conclusions When competing events exist, absolute risk is better estimated using competing risk regression, but etiologic associations by this method must be carefully interpreted. The presence of vascular disease in CKD decreases the likelihood of survival to ESRD, independent of age and other risk factors. PMID:22859747

Vascular disease, ESRD, and death: interpreting competing risk analyses.

PubMed

Grams, Morgan E; Coresh, Josef; Segev, Dorry L; Kucirka, Lauren M; Tighiouart, Hocine; Sarnak, Mark J

2012-10-01

Vascular disease, a common condition in CKD, is a risk factor for mortality and ESRD. Optimal patient care requires accurate estimation and ordering of these competing risks. This is a prospective cohort study of screened (n=885) and randomized participants (n=837) in the Modification of Diet in Renal Disease study (original study enrollment, 1989-1992), evaluating the association of vascular disease with ESRD and pre-ESRD mortality using standard survival analysis and competing risk regression. The method of analysis resulted in markedly different estimates. Cumulative incidence by standard analysis (censoring at the competing event) implied that, with vascular disease, the 15-year incidence was 66% and 51% for ESRD and pre-ESRD death, respectively. A more accurate representation of absolute risk was estimated with competing risk regression: 15-year incidence was 54% and 29% for ESRD and pre-ESRD death, respectively. For the association of vascular disease with pre-ESRD death, estimates of relative risk by the two methods were similar (standard survival analysis adjusted hazard ratio, 1.63; 95% confidence interval, 1.20-2.20; competing risk regression adjusted subhazard ratio, 1.57; 95% confidence interval, 1.15-2.14). In contrast, the hazard and subhazard ratios differed substantially for other associations, such as GFR and pre-ESRD mortality. When competing events exist, absolute risk is better estimated using competing risk regression, but etiologic associations by this method must be carefully interpreted. The presence of vascular disease in CKD decreases the likelihood of survival to ESRD, independent of age and other risk factors.

Univariate and bivariate likelihood-based meta-analysis methods performed comparably when marginal sensitivity and specificity were the targets of inference.

PubMed

Dahabreh, Issa J; Trikalinos, Thomas A; Lau, Joseph; Schmid, Christopher H

2017-03-01

To compare statistical methods for meta-analysis of sensitivity and specificity of medical tests (e.g., diagnostic or screening tests). We constructed a database of PubMed-indexed meta-analyses of test performance from which 2 × 2 tables for each included study could be extracted. We reanalyzed the data using univariate and bivariate random effects models fit with inverse variance and maximum likelihood methods. Analyses were performed using both normal and binomial likelihoods to describe within-study variability. The bivariate model using the binomial likelihood was also fit using a fully Bayesian approach. We use two worked examples-thoracic computerized tomography to detect aortic injury and rapid prescreening of Papanicolaou smears to detect cytological abnormalities-to highlight that different meta-analysis approaches can produce different results. We also present results from reanalysis of 308 meta-analyses of sensitivity and specificity. Models using the normal approximation produced sensitivity and specificity estimates closer to 50% and smaller standard errors compared to models using the binomial likelihood; absolute differences of 5% or greater were observed in 12% and 5% of meta-analyses for sensitivity and specificity, respectively. Results from univariate and bivariate random effects models were similar, regardless of estimation method. Maximum likelihood and Bayesian methods produced almost identical summary estimates under the bivariate model; however, Bayesian analyses indicated greater uncertainty around those estimates. Bivariate models produced imprecise estimates of the between-study correlation of sensitivity and specificity. Differences between methods were larger with increasing proportion of studies that were small or required a continuity correction. The binomial likelihood should be used to model within-study variability. Univariate and bivariate models give similar estimates of the marginal distributions for sensitivity and specificity. Bayesian methods fully quantify uncertainty and their ability to incorporate external evidence may be useful for imprecisely estimated parameters. Copyright © 2017 Elsevier Inc. All rights reserved.

Maximum likelihood estimation for Cox's regression model under nested case-control sampling.

PubMed

Scheike, Thomas H; Juul, Anders

2004-04-01

Nested case-control sampling is designed to reduce the costs of large cohort studies. It is important to estimate the parameters of interest as efficiently as possible. We present a new maximum likelihood estimator (MLE) for nested case-control sampling in the context of Cox's proportional hazards model. The MLE is computed by the EM-algorithm, which is easy to implement in the proportional hazards setting. Standard errors are estimated by a numerical profile likelihood approach based on EM aided differentiation. The work was motivated by a nested case-control study that hypothesized that insulin-like growth factor I was associated with ischemic heart disease. The study was based on a population of 3784 Danes and 231 cases of ischemic heart disease where controls were matched on age and gender. We illustrate the use of the MLE for these data and show how the maximum likelihood framework can be used to obtain information additional to the relative risk estimates of covariates.

Image informative maps for component-wise estimating parameters of signal-dependent noise

NASA Astrophysics Data System (ADS)

Uss, Mykhail L.; Vozel, Benoit; Lukin, Vladimir V.; Chehdi, Kacem

2013-01-01

We deal with the problem of blind parameter estimation of signal-dependent noise from mono-component image data. Multispectral or color images can be processed in a component-wise manner. The main results obtained rest on the assumption that the image texture and noise parameters estimation problems are interdependent. A two-dimensional fractal Brownian motion (fBm) model is used for locally describing image texture. A polynomial model is assumed for the purpose of describing the signal-dependent noise variance dependence on image intensity. Using the maximum likelihood approach, estimates of both fBm-model and noise parameters are obtained. It is demonstrated that Fisher information (FI) on noise parameters contained in an image is distributed nonuniformly over intensity coordinates (an image intensity range). It is also shown how to find the most informative intensities and the corresponding image areas for a given noisy image. The proposed estimator benefits from these detected areas to improve the estimation accuracy of signal-dependent noise parameters. Finally, the potential estimation accuracy (Cramér-Rao Lower Bound, or CRLB) of noise parameters is derived, providing confidence intervals of these estimates for a given image. In the experiment, the proposed and existing state-of-the-art noise variance estimators are compared for a large image database using CRLB-based statistical efficiency criteria.

How Affiliation Disclosure and Control Over User-Generated Comments Affects Consumer Health Knowledge and Behavior: A Randomized Controlled Experiment of Pharmaceutical Direct-to-Consumer Advertising on Social Media

PubMed Central

Vendemia, Megan Ashley

2016-01-01

Background More people are seeking health information online than ever before and pharmaceutical companies are increasingly marketing their drugs through social media. Objective The aim was to examine two major concerns related to online direct-to-consumer pharmaceutical advertising: (1) how disclosing an affiliation with a pharmaceutical company affects how people respond to drug information produced by both health organizations and online commenters, and (2) how knowledge that health organizations control the display of user-generated comments affects consumer health knowledge and behavior. Methods We conducted a 2×2×2 between-subjects experiment (N=674). All participants viewed an infographic posted to Facebook by a health organization about a prescription allergy drug. Across conditions, the infographic varied in the degree to which the health organization and commenters appeared to be affiliated with a drug manufacturer, and the display of user-generated comments appeared to be controlled. Results Affiliation disclosure statements on a health organization’s Facebook post increased perceptions of an organization-drug manufacturer connection, which reduced trust in the organization (point estimate –0.45, 95% CI –0.69 to –0.24) and other users who posted comments about the drug (point estimate –0.44, 95% CI –0.68 to –0.22). Furthermore, increased perceptions of an organization-manufacturer connection reduced the likelihood that people would recommend the drug to important others (point estimate –0.35, 95% CI –0.59 to –0.15), and share the drug post with others on Facebook (point estimate –0.37, 95% CI –0.64 to –0.16). An affiliation cue next to the commenters' names increased perceptions that the commenters were affiliated with the drug manufacturer, which reduced trust in the comments (point estimate –0.81, 95% CI –1.04 to –0.59), the organization that made the post (point estimate –0.68, 95% CI –0.90 to –0.49), the likelihood of participants recommending the drug (point estimate –0.61, 95% CI –0.82 to –0.43), and sharing the post with others on Facebook (point estimate –0.63, 95% CI –0.87 to –0.43). Cues indicating that a health organization removed user-generated comments from a post increased perceptions that the drug manufacturer influenced the display of the comments, which negatively affected trust in the comments (point estimate –0.35, 95% CI –0.53 to –0.20), the organization (point estimate –0.31, 95% CI –0.47 to –0.17), the likelihood of recommending the drug (point estimate –0.26, 95% CI –0.41 to –0.14), and the likelihood of sharing the post with others on Facebook (point estimate –0.28, 95% CI –0.45 to –0.15). (All estimates are unstandardized indirect effects and 95% bias-corrected bootstrap confidence intervals.) Conclusions Concern over pharmaceutical companies hiding their affiliations and strategically controlling user-generated comments is well founded; these practices can greatly affect not only how viewers evaluate drug information online, but also how likely they are to propagate the information throughout their online and offline social networks. PMID:27435883

A unified procedure for meta-analytic evaluation of surrogate end points in randomized clinical trials

PubMed Central

Dai, James Y.; Hughes, James P.

2012-01-01

The meta-analytic approach to evaluating surrogate end points assesses the predictiveness of treatment effect on the surrogate toward treatment effect on the clinical end point based on multiple clinical trials. Definition and estimation of the correlation of treatment effects were developed in linear mixed models and later extended to binary or failure time outcomes on a case-by-case basis. In a general regression setting that covers nonnormal outcomes, we discuss in this paper several metrics that are useful in the meta-analytic evaluation of surrogacy. We propose a unified 3-step procedure to assess these metrics in settings with binary end points, time-to-event outcomes, or repeated measures. First, the joint distribution of estimated treatment effects is ascertained by an estimating equation approach; second, the restricted maximum likelihood method is used to estimate the means and the variance components of the random treatment effects; finally, confidence intervals are constructed by a parametric bootstrap procedure. The proposed method is evaluated by simulations and applications to 2 clinical trials. PMID:22394448

On the use of secondary capture-recapture samples to estimate temporary emigration and breeding proportions

USGS Publications Warehouse

Kendall, W.L.; Nichols, J.D.; North, P.M.; Nichols, J.D.

1995-01-01

The use of the Cormack- Jolly-Seber model under a standard sampling scheme of one sample per time period, when the Jolly-Seber assumption that all emigration is permanent does not hold, leads to the confounding of temporary emigration probabilities with capture probabilities. This biases the estimates of capture probability when temporary emigration is a completely random process, and both capture and survival probabilities when there is a temporary trap response in temporary emigration, or it is Markovian. The use of secondary capture samples over a shorter interval within each period, during which the population is assumed to be closed (Pollock's robust design), provides a second source of information on capture probabilities. This solves the confounding problem, and thus temporary emigration probabilities can be estimated. This process can be accomplished in an ad hoc fashion for completely random temporary emigration and to some extent in the temporary trap response case, but modelling the complete sampling process provides more flexibility and permits direct estimation of variances. For the case of Markovian temporary emigration, a full likelihood is required.

Five Methods for Estimating Angoff Cut Scores with IRT

ERIC Educational Resources Information Center

Wyse, Adam E.

2017-01-01

This article illustrates five different methods for estimating Angoff cut scores using item response theory (IRT) models. These include maximum likelihood (ML), expected a priori (EAP), modal a priori (MAP), and weighted maximum likelihood (WML) estimators, as well as the most commonly used approach based on translating ratings through the test…

Computation of nonlinear least squares estimator and maximum likelihood using principles in matrix calculus

NASA Astrophysics Data System (ADS)

Mahaboob, B.; Venkateswarlu, B.; Sankar, J. Ravi; Balasiddamuni, P.

2017-11-01

This paper uses matrix calculus techniques to obtain Nonlinear Least Squares Estimator (NLSE), Maximum Likelihood Estimator (MLE) and Linear Pseudo model for nonlinear regression model. David Pollard and Peter Radchenko [1] explained analytic techniques to compute the NLSE. However the present research paper introduces an innovative method to compute the NLSE using principles in multivariate calculus. This study is concerned with very new optimization techniques used to compute MLE and NLSE. Anh [2] derived NLSE and MLE of a heteroscedatistic regression model. Lemcoff [3] discussed a procedure to get linear pseudo model for nonlinear regression model. In this research article a new technique is developed to get the linear pseudo model for nonlinear regression model using multivariate calculus. The linear pseudo model of Edmond Malinvaud [4] has been explained in a very different way in this paper. David Pollard et.al used empirical process techniques to study the asymptotic of the LSE (Least-squares estimation) for the fitting of nonlinear regression function in 2006. In Jae Myung [13] provided a go conceptual for Maximum likelihood estimation in his work “Tutorial on maximum likelihood estimation

SUPPLEMENT: “THE RATE OF BINARY BLACK HOLE MERGERS INFERRED FROM ADVANCED LIGO OBSERVATIONS SURROUNDING GW150914” (2016, ApJL, 833, L1)

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

Abbott, B. P.; Abbott, R.; Abernathy, M. R.

This article provides supplemental information for a Letter reporting the rate of (BBH) coalescences inferred from 16 days of coincident Advanced LIGO observations surrounding the transient (GW) signal GW150914. In that work we reported various rate estimates whose 90% confidence intervals fell in the range 2–600 Gpc{sup −3} yr{sup −1}. Here we give details on our method and computations, including information about our search pipelines, a derivation of our likelihood function for the analysis, a description of the astrophysical search trigger distribution expected from merging BBHs, details on our computational methods, a description of the effects and our model for calibration uncertainty,more » and an analytic method for estimating our detector sensitivity, which is calibrated to our measurements.« less

Physics-based, Bayesian sequential detection method and system for radioactive contraband

DOEpatents

Candy, James V; Axelrod, Michael C; Breitfeller, Eric F; Chambers, David H; Guidry, Brian L; Manatt, Douglas R; Meyer, Alan W; Sale, Kenneth E

2014-03-18

A distributed sequential method and system for detecting and identifying radioactive contraband from highly uncertain (noisy) low-count, radionuclide measurements, i.e. an event mode sequence (EMS), using a statistical approach based on Bayesian inference and physics-model-based signal processing based on the representation of a radionuclide as a monoenergetic decomposition of monoenergetic sources. For a given photon event of the EMS, the appropriate monoenergy processing channel is determined using a confidence interval condition-based discriminator for the energy amplitude and interarrival time and parameter estimates are used to update a measured probability density function estimate for a target radionuclide. A sequential likelihood ratio test is then used to determine one of two threshold conditions signifying that the EMS is either identified as the target radionuclide or not, and if not, then repeating the process for the next sequential photon event of the EMS until one of the two threshold conditions is satisfied.

Team climate, intention to leave and turnover among hospital employees: prospective cohort study.

PubMed

Kivimäki, Mika; Vanhala, Anna; Pentti, Jaana; Länsisalmi, Hannakaisa; Virtanen, Marianna; Elovainio, Marko; Vahtera, Jussi

2007-10-23

In hospitals, the costs of employee turnover are substantial and intentions to leave among staff may manifest as lowered performance. We examined whether team climate, as indicated by clear and shared goals, participation, task orientation and support for innovation, predicts intention to leave the job and actual turnover among hospital employees. Prospective study with baseline and follow-up surveys (2-4 years apart). The participants were 6,441 (785 men, 5,656 women) hospital employees under the age of 55 at the time of follow-up survey. Logistic regression with generalized estimating equations was used as an analysis method to include both individual and work unit level predictors in the models. Among stayers with no intention to leave at baseline, lower self-reported team climate predicted higher likelihood of having intentions to leave at follow-up (odds ratio per 1 standard deviation decrease in team climate was 1.6, 95% confidence interval 1.4-1.8). Lower co-worker assessed team climate at follow-up was also association with such intentions (odds ratio 1.8, 95% confidence interval 1.4-2.4). Among all participants, the likelihood of actually quitting the job was higher for those with poor self-reported team climate at baseline. This association disappeared after adjustment for intention to leave at baseline suggesting that such intentions may explain the greater turnover rate among employees with low team climate. Improving team climate may reduce intentions to leave and turnover among hospital employees.

The Statin-Associated Muscle Symptom Clinical Index (SAMS-CI): Revision for Clinical Use, Content Validation, and Inter-rater Reliability.

PubMed

Rosenson, Robert S; Miller, Kate; Bayliss, Martha; Sanchez, Robert J; Baccara-Dinet, Marie T; Chibedi-De-Roche, Daniela; Taylor, Beth; Khan, Irfan; Manvelian, Garen; White, Michelle; Jacobson, Terry A

2017-04-01

The Statin-Associated Muscle Symptom Clinical Index (SAMS-CI) is a method for assessing the likelihood that a patient's muscle symptoms (e.g., myalgia or myopathy) were caused or worsened by statin use. The objectives of this study were to prepare the SAMS-CI for clinical use, estimate its inter-rater reliability, and collect feedback from physicians on its practical application. For content validity, we conducted structured in-depth interviews with its original authors as well as with a panel of independent physicians. Estimation of inter-rater reliability involved an analysis of 30 written clinical cases which were scored by a sample of physicians. A separate group of physicians provided feedback on the clinical use of the SAMS-CI and its potential utility in practice. Qualitative interviews with providers supported the content validity of the SAMS-CI. Feedback on the clinical use of the SAMS-CI included several perceived benefits (such as brevity, clear wording, and simple scoring process) and some possible concerns (workflow issues and applicability in primary care). The inter-rater reliability of the SAMS-CI was estimated to be 0.77 (confidence interval 0.66-0.85), indicating high concordance between raters. With additional provider feedback, a revised SAMS-CI instrument was created suitable for further testing, both in the clinical setting and in prospective validation studies. With standardized questions, vetted language, easily interpreted scores, and demonstrated reliability, the SAMS aims to estimate the likelihood that a patient's muscle symptoms were attributable to statins. The SAMS-CI may support better detection of statin-associated muscle symptoms in clinical practice, optimize treatment for patients experiencing muscle symptoms, and provide a useful tool for further clinical research.

Estimating the Causal Impact of Proximity to Gold and Copper Mines on Respiratory Diseases in Chilean Children: An Application of Targeted Maximum Likelihood Estimation

PubMed Central

Herrera, Ronald; Berger, Ursula; von Ehrenstein, Ondine S.; Díaz, Iván; Huber, Stella; Moraga Muñoz, Daniel; Radon, Katja

2017-01-01

In a town located in a desert area of Northern Chile, gold and copper open-pit mining is carried out involving explosive processes. These processes are associated with increased dust exposure, which might affect children’s respiratory health. Therefore, we aimed to quantify the causal attributable risk of living close to the mines on asthma or allergic rhinoconjunctivitis risk burden in children. Data on the prevalence of respiratory diseases and potential confounders were available from a cross-sectional survey carried out in 2009 among 288 (response: 69%) children living in the community. The proximity of the children’s home addresses to the local gold and copper mine was calculated using geographical positioning systems. We applied targeted maximum likelihood estimation to obtain the causal attributable risk (CAR) for asthma, rhinoconjunctivitis and both outcomes combined. Children living more than the first quartile away from the mines were used as the unexposed group. Based on the estimated CAR, a hypothetical intervention in which all children lived at least one quartile away from the copper mine would decrease the risk of rhinoconjunctivitis by 4.7 percentage points (CAR: −4.7; 95% confidence interval (95% CI): −8.4; −0.11); and 4.2 percentage points (CAR: −4.2; 95% CI: −7.9;−0.05) for both outcomes combined. Overall, our results suggest that a hypothetical intervention intended to increase the distance between the place of residence of the highest exposed children would reduce the prevalence of respiratory disease in the community by around four percentage points. This approach could help local policymakers in the development of efficient public health strategies. PMID:29280971

Estimating the Causal Impact of Proximity to Gold and Copper Mines on Respiratory Diseases in Chilean Children: An Application of Targeted Maximum Likelihood Estimation.

PubMed

Herrera, Ronald; Berger, Ursula; von Ehrenstein, Ondine S; Díaz, Iván; Huber, Stella; Moraga Muñoz, Daniel; Radon, Katja

2017-12-27

In a town located in a desert area of Northern Chile, gold and copper open-pit mining is carried out involving explosive processes. These processes are associated with increased dust exposure, which might affect children's respiratory health. Therefore, we aimed to quantify the causal attributable risk of living close to the mines on asthma or allergic rhinoconjunctivitis risk burden in children. Data on the prevalence of respiratory diseases and potential confounders were available from a cross-sectional survey carried out in 2009 among 288 (response: 69 % ) children living in the community. The proximity of the children's home addresses to the local gold and copper mine was calculated using geographical positioning systems. We applied targeted maximum likelihood estimation to obtain the causal attributable risk (CAR) for asthma, rhinoconjunctivitis and both outcomes combined. Children living more than the first quartile away from the mines were used as the unexposed group. Based on the estimated CAR, a hypothetical intervention in which all children lived at least one quartile away from the copper mine would decrease the risk of rhinoconjunctivitis by 4.7 percentage points (CAR: - 4.7 ; 95 % confidence interval ( 95 % CI): - 8.4 ; - 0.11 ); and 4.2 percentage points (CAR: - 4.2 ; 95 % CI: - 7.9 ; - 0.05 ) for both outcomes combined. Overall, our results suggest that a hypothetical intervention intended to increase the distance between the place of residence of the highest exposed children would reduce the prevalence of respiratory disease in the community by around four percentage points. This approach could help local policymakers in the development of efficient public health strategies.

Models and analysis for multivariate failure time data

NASA Astrophysics Data System (ADS)

Shih, Joanna Huang

The goal of this research is to develop and investigate models and analytic methods for multivariate failure time data. We compare models in terms of direct modeling of the margins, flexibility of dependency structure, local vs. global measures of association, and ease of implementation. In particular, we study copula models, and models produced by right neutral cumulative hazard functions and right neutral hazard functions. We examine the changes of association over time for families of bivariate distributions induced from these models by displaying their density contour plots, conditional density plots, correlation curves of Doksum et al, and local cross ratios of Oakes. We know that bivariate distributions with same margins might exhibit quite different dependency structures. In addition to modeling, we study estimation procedures. For copula models, we investigate three estimation procedures. the first procedure is full maximum likelihood. The second procedure is two-stage maximum likelihood. At stage 1, we estimate the parameters in the margins by maximizing the marginal likelihood. At stage 2, we estimate the dependency structure by fixing the margins at the estimated ones. The third procedure is two-stage partially parametric maximum likelihood. It is similar to the second procedure, but we estimate the margins by the Kaplan-Meier estimate. We derive asymptotic properties for these three estimation procedures and compare their efficiency by Monte-Carlo simulations and direct computations. For models produced by right neutral cumulative hazards and right neutral hazards, we derive the likelihood and investigate the properties of the maximum likelihood estimates. Finally, we develop goodness of fit tests for the dependency structure in the copula models. We derive a test statistic and its asymptotic properties based on the test of homogeneity of Zelterman and Chen (1988), and a graphical diagnostic procedure based on the empirical Bayes approach. We study the performance of these two methods using actual and computer generated data.

Endoscopic ultrasound-guided fine needle core biopsy for the diagnosis of pancreatic malignant lesions: a systematic review and Meta-Analysis

PubMed Central

Yang, Yongtao; Li, Lianyong; Qu, Changmin; Liang, Shuwen; Zeng, Bolun; Luo, Zhiwen

2016-01-01

Endoscopic ultrasound-guided fine needle core biopsy (EUS-FNB) has been used as an effective method of diagnosing pancreatic malignant lesions. It has the advantage of providing well preserved tissue for histologic grading and subsequent molecular biological analysis. In order to estimate the diagnostic accuracy of EUS-FNB for pancreatic malignant lesions, studies assessing EUS-FNB to diagnose solid pancreatic masses were selected via Medline. Sixteen articles published between 2005 and 2015, covering 828 patients, met the inclusion criteria. The summary estimates for EUS-FNB differentiating malignant from benign solid pancreatic masses were: sensitivity 0.84 (95% confidence interval (CI), 0.82–0.87); specificity 0.98 (95% CI, 0.93–1.00); positive likelihood ratio 8.0 (95% CI 4.5–14.4); negative likelihood ratio 0.17 (95% CI 0.10–0.26); and DOR 64 (95% CI 30.4–134.8). The area under the sROC curve was 0.96. Subgroup analysis did not identify other factors that could substantially affect the diagnostic accuracy, such as the study design, location of study, number of centers, location of lesion, whether or not a cytopathologist was present, and so on. EUS-FNB is a reliable diagnostic tool for solid pancreatic masses and should be especially considered for pathology where histologic morphology is preferred for diagnosis. PMID:26960914

Explaining the effect of event valence on unrealistic optimism.

PubMed

Gold, Ron S; Brown, Mark G

2009-05-01

People typically exhibit 'unrealistic optimism' (UO): they believe they have a lower chance of experiencing negative events and a higher chance of experiencing positive events than does the average person. UO has been found to be greater for negative than positive events. This 'valence effect' has been explained in terms of motivational processes. An alternative explanation is provided by the 'numerosity model', which views the valence effect simply as a by-product of a tendency for likelihood estimates pertaining to the average member of a group to increase with the size of the group. Predictions made by the numerosity model were tested in two studies. In each, UO for a single event was assessed. In Study 1 (n = 115 students), valence was manipulated by framing the event either negatively or positively, and participants estimated their own likelihood and that of the average student at their university. In Study 2 (n = 139 students), valence was again manipulated and participants again estimated their own likelihood; additionally, group size was manipulated by having participants estimate the likelihood of the average student in a small, medium-sized, or large group. In each study, the valence effect was found, but was due to an effect on estimates of own likelihood, not the average person's likelihood. In Study 2, valence did not interact with group size. The findings contradict the numerosity model, but are in accord with the motivational explanation. Implications for health education are discussed.

A new framework of statistical inferences based on the valid joint sampling distribution of the observed counts in an incomplete contingency table.

PubMed

Tian, Guo-Liang; Li, Hui-Qiong

2017-08-01

Some existing confidence interval methods and hypothesis testing methods in the analysis of a contingency table with incomplete observations in both margins entirely depend on an underlying assumption that the sampling distribution of the observed counts is a product of independent multinomial/binomial distributions for complete and incomplete counts. However, it can be shown that this independency assumption is incorrect and can result in unreliable conclusions because of the under-estimation of the uncertainty. Therefore, the first objective of this paper is to derive the valid joint sampling distribution of the observed counts in a contingency table with incomplete observations in both margins. The second objective is to provide a new framework for analyzing incomplete contingency tables based on the derived joint sampling distribution of the observed counts by developing a Fisher scoring algorithm to calculate maximum likelihood estimates of parameters of interest, the bootstrap confidence interval methods, and the bootstrap testing hypothesis methods. We compare the differences between the valid sampling distribution and the sampling distribution under the independency assumption. Simulation studies showed that average/expected confidence-interval widths of parameters based on the sampling distribution under the independency assumption are shorter than those based on the new sampling distribution, yielding unrealistic results. A real data set is analyzed to illustrate the application of the new sampling distribution for incomplete contingency tables and the analysis results again confirm the conclusions obtained from the simulation studies.

Genealogical Working Distributions for Bayesian Model Testing with Phylogenetic Uncertainty

PubMed Central

Baele, Guy; Lemey, Philippe; Suchard, Marc A.

2016-01-01

Marginal likelihood estimates to compare models using Bayes factors frequently accompany Bayesian phylogenetic inference. Approaches to estimate marginal likelihoods have garnered increased attention over the past decade. In particular, the introduction of path sampling (PS) and stepping-stone sampling (SS) into Bayesian phylogenetics has tremendously improved the accuracy of model selection. These sampling techniques are now used to evaluate complex evolutionary and population genetic models on empirical data sets, but considerable computational demands hamper their widespread adoption. Further, when very diffuse, but proper priors are specified for model parameters, numerical issues complicate the exploration of the priors, a necessary step in marginal likelihood estimation using PS or SS. To avoid such instabilities, generalized SS (GSS) has recently been proposed, introducing the concept of “working distributions” to facilitate—or shorten—the integration process that underlies marginal likelihood estimation. However, the need to fix the tree topology currently limits GSS in a coalescent-based framework. Here, we extend GSS by relaxing the fixed underlying tree topology assumption. To this purpose, we introduce a “working” distribution on the space of genealogies, which enables estimating marginal likelihoods while accommodating phylogenetic uncertainty. We propose two different “working” distributions that help GSS to outperform PS and SS in terms of accuracy when comparing demographic and evolutionary models applied to synthetic data and real-world examples. Further, we show that the use of very diffuse priors can lead to a considerable overestimation in marginal likelihood when using PS and SS, while still retrieving the correct marginal likelihood using both GSS approaches. The methods used in this article are available in BEAST, a powerful user-friendly software package to perform Bayesian evolutionary analyses. PMID:26526428

Modelling the initial phase of an epidemic using incidence and infection network data: 2009 H1N1 pandemic in Israel as a case study

PubMed Central

Katriel, G.; Yaari, R.; Huppert, A.; Roll, U.; Stone, L.

2011-01-01

This paper presents new computational and modelling tools for studying the dynamics of an epidemic in its initial stages that use both available incidence time series and data describing the population's infection network structure. The work is motivated by data collected at the beginning of the H1N1 pandemic outbreak in Israel in the summer of 2009. We formulated a new discrete-time stochastic epidemic SIR (susceptible-infected-recovered) model that explicitly takes into account the disease's specific generation-time distribution and the intrinsic demographic stochasticity inherent to the infection process. Moreover, in contrast with many other modelling approaches, the model allows direct analytical derivation of estimates for the effective reproductive number (Re) and of their credible intervals, by maximum likelihood and Bayesian methods. The basic model can be extended to include age–class structure, and a maximum likelihood methodology allows us to estimate the model's next-generation matrix by combining two types of data: (i) the incidence series of each age group, and (ii) infection network data that provide partial information of ‘who-infected-who’. Unlike other approaches for estimating the next-generation matrix, the method developed here does not require making a priori assumptions about the structure of the next-generation matrix. We show, using a simulation study, that even a relatively small amount of information about the infection network greatly improves the accuracy of estimation of the next-generation matrix. The method is applied in practice to estimate the next-generation matrix from the Israeli H1N1 pandemic data. The tools developed here should be of practical importance for future investigations of epidemics during their initial stages. However, they require the availability of data which represent a random sample of the real epidemic process. We discuss the conditions under which reporting rates may or may not influence our estimated quantities and the effects of bias. PMID:21247949

Estimating a Logistic Discrimination Functions When One of the Training Samples Is Subject to Misclassification: A Maximum Likelihood Approach.

PubMed

Nagelkerke, Nico; Fidler, Vaclav

2015-01-01

The problem of discrimination and classification is central to much of epidemiology. Here we consider the estimation of a logistic regression/discrimination function from training samples, when one of the training samples is subject to misclassification or mislabeling, e.g. diseased individuals are incorrectly classified/labeled as healthy controls. We show that this leads to zero-inflated binomial model with a defective logistic regression or discrimination function, whose parameters can be estimated using standard statistical methods such as maximum likelihood. These parameters can be used to estimate the probability of true group membership among those, possibly erroneously, classified as controls. Two examples are analyzed and discussed. A simulation study explores properties of the maximum likelihood parameter estimates and the estimates of the number of mislabeled observations.

Parameter estimation in astronomy through application of the likelihood ratio. [satellite data analysis techniques

NASA Technical Reports Server (NTRS)

Cash, W.

1979-01-01

Many problems in the experimental estimation of parameters for models can be solved through use of the likelihood ratio test. Applications of the likelihood ratio, with particular attention to photon counting experiments, are discussed. The procedures presented solve a greater range of problems than those currently in use, yet are no more difficult to apply. The procedures are proved analytically, and examples from current problems in astronomy are discussed.

Cosmic shear measurement with maximum likelihood and maximum a posteriori inference

NASA Astrophysics Data System (ADS)

Hall, Alex; Taylor, Andy

2017-06-01

We investigate the problem of noise bias in maximum likelihood and maximum a posteriori estimators for cosmic shear. We derive the leading and next-to-leading order biases and compute them in the context of galaxy ellipticity measurements, extending previous work on maximum likelihood inference for weak lensing. We show that a large part of the bias on these point estimators can be removed using information already contained in the likelihood when a galaxy model is specified, without the need for external calibration. We test these bias-corrected estimators on simulated galaxy images similar to those expected from planned space-based weak lensing surveys, with promising results. We find that the introduction of an intrinsic shape prior can help with mitigation of noise bias, such that the maximum a posteriori estimate can be made less biased than the maximum likelihood estimate. Second-order terms offer a check on the convergence of the estimators, but are largely subdominant. We show how biases propagate to shear estimates, demonstrating in our simple set-up that shear biases can be reduced by orders of magnitude and potentially to within the requirements of planned space-based surveys at mild signal-to-noise ratio. We find that second-order terms can exhibit significant cancellations at low signal-to-noise ratio when Gaussian noise is assumed, which has implications for inferring the performance of shear-measurement algorithms from simplified simulations. We discuss the viability of our point estimators as tools for lensing inference, arguing that they allow for the robust measurement of ellipticity and shear.

A Comparison of Pseudo-Maximum Likelihood and Asymptotically Distribution-Free Dynamic Factor Analysis Parameter Estimation in Fitting Covariance Structure Models to Block-Toeplitz Matrices Representing Single-Subject Multivariate Time-Series.

ERIC Educational Resources Information Center

Molenaar, Peter C. M.; Nesselroade, John R.

1998-01-01

Pseudo-Maximum Likelihood (p-ML) and Asymptotically Distribution Free (ADF) estimation methods for estimating dynamic factor model parameters within a covariance structure framework were compared through a Monte Carlo simulation. Both methods appear to give consistent model parameter estimates, but only ADF gives standard errors and chi-square…

Log-normal frailty models fitted as Poisson generalized linear mixed models.

PubMed

Hirsch, Katharina; Wienke, Andreas; Kuss, Oliver

2016-12-01

The equivalence of a survival model with a piecewise constant baseline hazard function and a Poisson regression model has been known since decades. As shown in recent studies, this equivalence carries over to clustered survival data: A frailty model with a log-normal frailty term can be interpreted and estimated as a generalized linear mixed model with a binary response, a Poisson likelihood, and a specific offset. Proceeding this way, statistical theory and software for generalized linear mixed models are readily available for fitting frailty models. This gain in flexibility comes at the small price of (1) having to fix the number of pieces for the baseline hazard in advance and (2) having to "explode" the data set by the number of pieces. In this paper we extend the simulations of former studies by using a more realistic baseline hazard (Gompertz) and by comparing the model under consideration with competing models. Furthermore, the SAS macro %PCFrailty is introduced to apply the Poisson generalized linear mixed approach to frailty models. The simulations show good results for the shared frailty model. Our new %PCFrailty macro provides proper estimates, especially in case of 4 events per piece. The suggested Poisson generalized linear mixed approach for log-normal frailty models based on the %PCFrailty macro provides several advantages in the analysis of clustered survival data with respect to more flexible modelling of fixed and random effects, exact (in the sense of non-approximate) maximum likelihood estimation, and standard errors and different types of confidence intervals for all variance parameters. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

Cosmological parameters from a re-analysis of the WMAP 7 year low-resolution maps

NASA Astrophysics Data System (ADS)

Finelli, F.; De Rosa, A.; Gruppuso, A.; Paoletti, D.

2013-06-01

Cosmological parameters from Wilkinson Microwave Anisotropy Probe (WMAP) 7 year data are re-analysed by substituting a pixel-based likelihood estimator to the one delivered publicly by the WMAP team. Our pixel-based estimator handles exactly intensity and polarization in a joint manner, allowing us to use low-resolution maps and noise covariance matrices in T, Q, U at the same resolution, which in this work is 3.6°. We describe the features and the performances of the code implementing our pixel-based likelihood estimator. We perform a battery of tests on the application of our pixel-based likelihood routine to WMAP publicly available low-resolution foreground-cleaned products, in combination with the WMAP high-ℓ likelihood, reporting the differences on cosmological parameters evaluated by the full WMAP likelihood public package. The differences are not only due to the treatment of polarization, but also to the marginalization over monopole and dipole uncertainties present in the WMAP pixel likelihood code for temperature. The credible central value for the cosmological parameters change below the 1σ level with respect to the evaluation by the full WMAP 7 year likelihood code, with the largest difference in a shift to smaller values of the scalar spectral index nS.

N-mixture models for estimating population size from spatially replicated counts

USGS Publications Warehouse

Royle, J. Andrew

2004-01-01

Spatial replication is a common theme in count surveys of animals. Such surveys often generate sparse count data from which it is difficult to estimate population size while formally accounting for detection probability. In this article, i describe a class of models (n-mixture models) which allow for estimation of population size from such data. The key idea is to view site-specific population sizes, n, as independent random variables distributed according to some mixing distribution (e.g., Poisson). Prior parameters are estimated from the marginal likelihood of the data, having integrated over the prior distribution for n. Carroll and lombard (1985, journal of american statistical association 80, 423-426) proposed a class of estimators based on mixing over a prior distribution for detection probability. Their estimator can be applied in limited settings, but is sensitive to prior parameter values that are fixed a priori. Spatial replication provides additional information regarding the parameters of the prior distribution on n that is exploited by the n-mixture models and which leads to reasonable estimates of abundance from sparse data. A simulation study demonstrates superior operating characteristics (bias, confidence interval coverage) of the n-mixture estimator compared to the caroll and lombard estimator. Both estimators are applied to point count data on six species of birds illustrating the sensitivity to choice of prior on p and substantially different estimates of abundance as a consequence.

Performing Contrast Analysis in Factorial Designs: From NHST to Confidence Intervals and Beyond

PubMed Central

Wiens, Stefan; Nilsson, Mats E.

2016-01-01

Because of the continuing debates about statistics, many researchers may feel confused about how to analyze and interpret data. Current guidelines in psychology advocate the use of effect sizes and confidence intervals (CIs). However, researchers may be unsure about how to extract effect sizes from factorial designs. Contrast analysis is helpful because it can be used to test specific questions of central interest in studies with factorial designs. It weighs several means and combines them into one or two sets that can be tested with t tests. The effect size produced by a contrast analysis is simply the difference between means. The CI of the effect size informs directly about direction, hypothesis exclusion, and the relevance of the effects of interest. However, any interpretation in terms of precision or likelihood requires the use of likelihood intervals or credible intervals (Bayesian). These various intervals and even a Bayesian t test can be obtained easily with free software. This tutorial reviews these methods to guide researchers in answering the following questions: When I analyze mean differences in factorial designs, where can I find the effects of central interest, and what can I learn about their effect sizes? PMID:29805179

Procedure for estimating stability and control parameters from flight test data by using maximum likelihood methods employing a real-time digital system

NASA Technical Reports Server (NTRS)

Grove, R. D.; Bowles, R. L.; Mayhew, S. C.

1972-01-01

A maximum likelihood parameter estimation procedure and program were developed for the extraction of the stability and control derivatives of aircraft from flight test data. Nonlinear six-degree-of-freedom equations describing aircraft dynamics were used to derive sensitivity equations for quasilinearization. The maximum likelihood function with quasilinearization was used to derive the parameter change equations, the covariance matrices for the parameters and measurement noise, and the performance index function. The maximum likelihood estimator was mechanized into an iterative estimation procedure utilizing a real time digital computer and graphic display system. This program was developed for 8 measured state variables and 40 parameters. Test cases were conducted with simulated data for validation of the estimation procedure and program. The program was applied to a V/STOL tilt wing aircraft, a military fighter airplane, and a light single engine airplane. The particular nonlinear equations of motion, derivation of the sensitivity equations, addition of accelerations into the algorithm, operational features of the real time digital system, and test cases are described.

Modeling of 2D diffusion processes based on microscopy data: parameter estimation and practical identifiability analysis.

PubMed

Hock, Sabrina; Hasenauer, Jan; Theis, Fabian J

2013-01-01

Diffusion is a key component of many biological processes such as chemotaxis, developmental differentiation and tissue morphogenesis. Since recently, the spatial gradients caused by diffusion can be assessed in-vitro and in-vivo using microscopy based imaging techniques. The resulting time-series of two dimensional, high-resolutions images in combination with mechanistic models enable the quantitative analysis of the underlying mechanisms. However, such a model-based analysis is still challenging due to measurement noise and sparse observations, which result in uncertainties of the model parameters. We introduce a likelihood function for image-based measurements with log-normal distributed noise. Based upon this likelihood function we formulate the maximum likelihood estimation problem, which is solved using PDE-constrained optimization methods. To assess the uncertainty and practical identifiability of the parameters we introduce profile likelihoods for diffusion processes. As proof of concept, we model certain aspects of the guidance of dendritic cells towards lymphatic vessels, an example for haptotaxis. Using a realistic set of artificial measurement data, we estimate the five kinetic parameters of this model and compute profile likelihoods. Our novel approach for the estimation of model parameters from image data as well as the proposed identifiability analysis approach is widely applicable to diffusion processes. The profile likelihood based method provides more rigorous uncertainty bounds in contrast to local approximation methods.

Prediction of posttraumatic stress disorder symptomatology after childbirth - A Croatian longitudinal study.

PubMed

Srkalović Imširagić, Azijada; Begić, Dražen; Šimičević, Livija; Bajić, Žarko

2017-02-01

Following childbirth, a vast number of women experience some degree of mood swings, while some experience symptoms of postpartum posttraumatic stress disorder. Using a biopsychosocial model, the primary aim of this study was to identify predictors of posttraumatic stress disorder and its symptomatology following childbirth. This observational, longitudinal study included 372 postpartum women. In order to explore biopsychosocial predictors, participants completed several questionnaires 3-5 days after childbirth: the Impact of Events Scale Revised, the Big Five Inventory, The Edinburgh Postnatal Depression Scale, breastfeeding practice and social and demographic factors. Six to nine weeks after childbirth, participants re-completed the questionnaires regarding psychiatric symptomatology and breastfeeding practice. Using a multivariate level of analysis, the predictors that increased the likelihood of postpartum posttraumatic stress disorder symptomatology at the first study phase were: emergency caesarean section (odds ratio 2.48; confidence interval 1.13-5.43) and neuroticism personality trait (odds ratio 1.12; confidence interval 1.05-1.20). The predictor that increased the likelihood of posttraumatic stress disorder symptomatology at the second study phase was the baseline Impact of Events Scale Revised score (odds ratio 12.55; confidence interval 4.06-38.81). Predictors that decreased the likelihood of symptomatology at the second study phase were life in a nuclear family (odds ratio 0.27; confidence interval 0.09-0.77) and life in a city (odds ratio 0.29; confidence interval 0.09-0.94). Biopsychosocial theory is applicable to postpartum psychiatric disorders. In addition to screening for depression amongst postpartum women, there is a need to include other postpartum psychiatric symptomatology screenings in routine practice. Copyright © 2016 Australian College of Midwives. Published by Elsevier Ltd. All rights reserved.

On the Log-Normality of Historical Magnetic-Storm Intensity Statistics: Implications for Extreme-Event Probabilities

NASA Astrophysics Data System (ADS)

Love, J. J.; Rigler, E. J.; Pulkkinen, A. A.; Riley, P.

2015-12-01

An examination is made of the hypothesis that the statistics of magnetic-storm-maximum intensities are the realization of a log-normal stochastic process. Weighted least-squares and maximum-likelihood methods are used to fit log-normal functions to -Dst storm-time maxima for years 1957-2012; bootstrap analysis is used to established confidence limits on forecasts. Both methods provide fits that are reasonably consistent with the data; both methods also provide fits that are superior to those that can be made with a power-law function. In general, the maximum-likelihood method provides forecasts having tighter confidence intervals than those provided by weighted least-squares. From extrapolation of maximum-likelihood fits: a magnetic storm with intensity exceeding that of the 1859 Carrington event, -Dst > 850 nT, occurs about 1.13 times per century and a wide 95% confidence interval of [0.42, 2.41] times per century; a 100-yr magnetic storm is identified as having a -Dst > 880 nT (greater than Carrington) but a wide 95% confidence interval of [490, 1187] nT. This work is partially motivated by United States National Science and Technology Council and Committee on Space Research and International Living with a Star priorities and strategic plans for the assessment and mitigation of space-weather hazards.

Some Small Sample Results for Maximum Likelihood Estimation in Multidimensional Scaling.

ERIC Educational Resources Information Center

Ramsay, J. O.

1980-01-01

Some aspects of the small sample behavior of maximum likelihood estimates in multidimensional scaling are investigated with Monte Carlo techniques. In particular, the chi square test for dimensionality is examined and a correction for bias is proposed and evaluated. (Author/JKS)

A spatially explicit capture-recapture estimator for single-catch traps.

PubMed

Distiller, Greg; Borchers, David L

2015-11-01

Single-catch traps are frequently used in live-trapping studies of small mammals. Thus far, a likelihood for single-catch traps has proven elusive and usually the likelihood for multicatch traps is used for spatially explicit capture-recapture (SECR) analyses of such data. Previous work found the multicatch likelihood to provide a robust estimator of average density. We build on a recently developed continuous-time model for SECR to derive a likelihood for single-catch traps. We use this to develop an estimator based on observed capture times and compare its performance by simulation to that of the multicatch estimator for various scenarios with nonconstant density surfaces. While the multicatch estimator is found to be a surprisingly robust estimator of average density, its performance deteriorates with high trap saturation and increasing density gradients. Moreover, it is found to be a poor estimator of the height of the detection function. By contrast, the single-catch estimators of density, distribution, and detection function parameters are found to be unbiased or nearly unbiased in all scenarios considered. This gain comes at the cost of higher variance. If there is no interest in interpreting the detection function parameters themselves, and if density is expected to be fairly constant over the survey region, then the multicatch estimator performs well with single-catch traps. However if accurate estimation of the detection function is of interest, or if density is expected to vary substantially in space, then there is merit in using the single-catch estimator when trap saturation is above about 60%. The estimator's performance is improved if care is taken to place traps so as to span the range of variables that affect animal distribution. As a single-catch likelihood with unknown capture times remains intractable for now, researchers using single-catch traps should aim to incorporate timing devices with their traps.

Maintained Individual Data Distributed Likelihood Estimation (MIDDLE)

PubMed Central

Boker, Steven M.; Brick, Timothy R.; Pritikin, Joshua N.; Wang, Yang; von Oertzen, Timo; Brown, Donald; Lach, John; Estabrook, Ryne; Hunter, Michael D.; Maes, Hermine H.; Neale, Michael C.

2015-01-01

Maintained Individual Data Distributed Likelihood Estimation (MIDDLE) is a novel paradigm for research in the behavioral, social, and health sciences. The MIDDLE approach is based on the seemingly-impossible idea that data can be privately maintained by participants and never revealed to researchers, while still enabling statistical models to be fit and scientific hypotheses tested. MIDDLE rests on the assumption that participant data should belong to, be controlled by, and remain in the possession of the participants themselves. Distributed likelihood estimation refers to fitting statistical models by sending an objective function and vector of parameters to each participants’ personal device (e.g., smartphone, tablet, computer), where the likelihood of that individual’s data is calculated locally. Only the likelihood value is returned to the central optimizer. The optimizer aggregates likelihood values from responding participants and chooses new vectors of parameters until the model converges. A MIDDLE study provides significantly greater privacy for participants, automatic management of opt-in and opt-out consent, lower cost for the researcher and funding institute, and faster determination of results. Furthermore, if a participant opts into several studies simultaneously and opts into data sharing, these studies automatically have access to individual-level longitudinal data linked across all studies. PMID:26717128

Generalized weighted likelihood density estimators with application to finite mixture of exponential family distributions

PubMed Central

Zhan, Tingting; Chevoneva, Inna; Iglewicz, Boris

2010-01-01

The family of weighted likelihood estimators largely overlaps with minimum divergence estimators. They are robust to data contaminations compared to MLE. We define the class of generalized weighted likelihood estimators (GWLE), provide its influence function and discuss the efficiency requirements. We introduce a new truncated cubic-inverse weight, which is both first and second order efficient and more robust than previously reported weights. We also discuss new ways of selecting the smoothing bandwidth and weighted starting values for the iterative algorithm. The advantage of the truncated cubic-inverse weight is illustrated in a simulation study of three-components normal mixtures model with large overlaps and heavy contaminations. A real data example is also provided. PMID:20835375

Bayesian structural equation modeling in sport and exercise psychology.

PubMed

Stenling, Andreas; Ivarsson, Andreas; Johnson, Urban; Lindwall, Magnus

2015-08-01

Bayesian statistics is on the rise in mainstream psychology, but applications in sport and exercise psychology research are scarce. In this article, the foundations of Bayesian analysis are introduced, and we will illustrate how to apply Bayesian structural equation modeling in a sport and exercise psychology setting. More specifically, we contrasted a confirmatory factor analysis on the Sport Motivation Scale II estimated with the most commonly used estimator, maximum likelihood, and a Bayesian approach with weakly informative priors for cross-loadings and correlated residuals. The results indicated that the model with Bayesian estimation and weakly informative priors provided a good fit to the data, whereas the model estimated with a maximum likelihood estimator did not produce a well-fitting model. The reasons for this discrepancy between maximum likelihood and Bayesian estimation are discussed as well as potential advantages and caveats with the Bayesian approach.

New estimates of the CMB angular power spectra from the WMAP 5 year low-resolution data

NASA Astrophysics Data System (ADS)

Gruppuso, A.; de Rosa, A.; Cabella, P.; Paci, F.; Finelli, F.; Natoli, P.; de Gasperis, G.; Mandolesi, N.

2009-11-01

A quadratic maximum likelihood (QML) estimator is applied to the Wilkinson Microwave Anisotropy Probe (WMAP) 5 year low-resolution maps to compute the cosmic microwave background angular power spectra (APS) at large scales for both temperature and polarization. Estimates and error bars for the six APS are provided up to l = 32 and compared, when possible, to those obtained by the WMAP team, without finding any inconsistency. The conditional likelihood slices are also computed for the Cl of all the six power spectra from l = 2 to 10 through a pixel-based likelihood code. Both the codes treat the covariance for (T, Q, U) in a single matrix without employing any approximation. The inputs of both the codes (foreground-reduced maps, related covariances and masks) are provided by the WMAP team. The peaks of the likelihood slices are always consistent with the QML estimates within the error bars; however, an excellent agreement occurs when the QML estimates are used as a fiducial power spectrum instead of the best-fitting theoretical power spectrum. By the full computation of the conditional likelihood on the estimated spectra, the value of the temperature quadrupole CTTl=2 is found to be less than 2σ away from the WMAP 5 year Λ cold dark matter best-fitting value. The BB spectrum is found to be well consistent with zero, and upper limits on the B modes are provided. The parity odd signals TB and EB are found to be consistent with zero.

The role of temporal call structure in species recognition of male Allobates talamancae (Cope, 1875): (Anura: Dendrobatidae).

PubMed

Kollarits, Dennis; Wappl, Christian; Ringler, Max

2017-01-30

Acoustic species recognition in anurans depends on spectral and temporal characteristics of the advertisement call. The recognition space of a species is shaped by the likelihood of heterospecific acoustic interference. The dendrobatid frogs Allobates talamancae (Cope, 1875) and Silverstoneia flotator (Dunn, 1931) occur syntopically in south-west Costa Rica. A previous study showed that these two species avoid acoustic interference by spectral stratification. In this study, the role of the temporal call structure in the advertisement call of A. talamancae was analyzed, in particular the internote-interval duration in providing species specific temporal cues. In playback trials, artificial advertisement calls with internote-intervals deviating up to ± 90 % from the population mean internote-interval were broadcast to vocally active territorial males. The phonotactic reactions of the males indicated that, unlike in closely related species, internote-interval duration is not a call property essential for species recognition in A. talamancae . However, temporal call structure may be used for species recognition when the likelihood of heterospecific interference is high. Also, the close-encounter courtship call of male A. talamancae is described.

Computing maximum-likelihood estimates for parameters of the National Descriptive Model of Mercury in Fish

USGS Publications Warehouse

Donato, David I.

2012-01-01

This report presents the mathematical expressions and the computational techniques required to compute maximum-likelihood estimates for the parameters of the National Descriptive Model of Mercury in Fish (NDMMF), a statistical model used to predict the concentration of methylmercury in fish tissue. The expressions and techniques reported here were prepared to support the development of custom software capable of computing NDMMF parameter estimates more quickly and using less computer memory than is currently possible with available general-purpose statistical software. Computation of maximum-likelihood estimates for the NDMMF by numerical solution of a system of simultaneous equations through repeated Newton-Raphson iterations is described. This report explains the derivation of the mathematical expressions required for computational parameter estimation in sufficient detail to facilitate future derivations for any revised versions of the NDMMF that may be developed.

A New Insight into the Earthquake Recurrence Studies from the Three-parameter Generalized Exponential Distributions

NASA Astrophysics Data System (ADS)

Pasari, S.; Kundu, D.; Dikshit, O.

2012-12-01

Earthquake recurrence interval is one of the important ingredients towards probabilistic seismic hazard assessment (PSHA) for any location. Exponential, gamma, Weibull and lognormal distributions are quite established probability models in this recurrence interval estimation. However, they have certain shortcomings too. Thus, it is imperative to search for some alternative sophisticated distributions. In this paper, we introduce a three-parameter (location, scale and shape) exponentiated exponential distribution and investigate the scope of this distribution as an alternative of the afore-mentioned distributions in earthquake recurrence studies. This distribution is a particular member of the exponentiated Weibull distribution. Despite of its complicated form, it is widely accepted in medical and biological applications. Furthermore, it shares many physical properties with gamma and Weibull family. Unlike gamma distribution, the hazard function of generalized exponential distribution can be easily computed even if the shape parameter is not an integer. To contemplate the plausibility of this model, a complete and homogeneous earthquake catalogue of 20 events (M ≥ 7.0) spanning for the period 1846 to 1995 from North-East Himalayan region (20-32 deg N and 87-100 deg E) has been used. The model parameters are estimated using maximum likelihood estimator (MLE) and method of moment estimator (MOME). No geological or geophysical evidences have been considered in this calculation. The estimated conditional probability reaches quite high after about a decade for an elapsed time of 17 years (i.e. 2012). Moreover, this study shows that the generalized exponential distribution fits the above data events more closely compared to the conventional models and hence it is tentatively concluded that generalized exponential distribution can be effectively considered in earthquake recurrence studies.

Maximum Likelihood Estimation of Nonlinear Structural Equation Models.

ERIC Educational Resources Information Center

Lee, Sik-Yum; Zhu, Hong-Tu

2002-01-01

Developed an EM type algorithm for maximum likelihood estimation of a general nonlinear structural equation model in which the E-step is completed by a Metropolis-Hastings algorithm. Illustrated the methodology with results from a simulation study and two real examples using data from previous studies. (SLD)

A New Monte Carlo Method for Estimating Marginal Likelihoods.

PubMed

Wang, Yu-Bo; Chen, Ming-Hui; Kuo, Lynn; Lewis, Paul O

2018-06-01

Evaluating the marginal likelihood in Bayesian analysis is essential for model selection. Estimators based on a single Markov chain Monte Carlo sample from the posterior distribution include the harmonic mean estimator and the inflated density ratio estimator. We propose a new class of Monte Carlo estimators based on this single Markov chain Monte Carlo sample. This class can be thought of as a generalization of the harmonic mean and inflated density ratio estimators using a partition weighted kernel (likelihood times prior). We show that our estimator is consistent and has better theoretical properties than the harmonic mean and inflated density ratio estimators. In addition, we provide guidelines on choosing optimal weights. Simulation studies were conducted to examine the empirical performance of the proposed estimator. We further demonstrate the desirable features of the proposed estimator with two real data sets: one is from a prostate cancer study using an ordinal probit regression model with latent variables; the other is for the power prior construction from two Eastern Cooperative Oncology Group phase III clinical trials using the cure rate survival model with similar objectives.

Analysis of photon count data from single-molecule fluorescence experiments

NASA Astrophysics Data System (ADS)

Burzykowski, T.; Szubiakowski, J.; Rydén, T.

2003-03-01

We consider single-molecule fluorescence experiments with data in the form of counts of photons registered over multiple time-intervals. Based on the observation schemes, linking back to works by Dehmelt [Bull. Am. Phys. Soc. 20 (1975) 60] and Cook and Kimble [Phys. Rev. Lett. 54 (1985) 1023], we propose an analytical approach to the data based on the theory of Markov-modulated Poisson processes (MMPP). In particular, we consider maximum-likelihood estimation. The method is illustrated using a real-life dataset. Additionally, the properties of the proposed method are investigated through simulations and compared to two other approaches developed by Yip et al. [J. Phys. Chem. A 102 (1998) 7564] and Molski [Chem. Phys. Lett. 324 (2000) 301].

Statistical analysis of field data for aircraft warranties

NASA Astrophysics Data System (ADS)

Lakey, Mary J.

Air Force and Navy maintenance data collection systems were researched to determine their scientific applicability to the warranty process. New and unique algorithms were developed to extract failure distributions which were then used to characterize how selected families of equipment typically fails. Families of similar equipment were identified in terms of function, technology and failure patterns. Statistical analyses and applications such as goodness-of-fit test, maximum likelihood estimation and derivation of confidence intervals for the probability density function parameters were applied to characterize the distributions and their failure patterns. Statistical and reliability theory, with relevance to equipment design and operational failures were also determining factors in characterizing the failure patterns of the equipment families. Inferences about the families with relevance to warranty needs were then made.

A Semiparametric Approach for Composite Functional Mapping of Dynamic Quantitative Traits

PubMed Central

Yang, Runqing; Gao, Huijiang; Wang, Xin; Zhang, Ji; Zeng, Zhao-Bang; Wu, Rongling

2007-01-01

Functional mapping has emerged as a powerful tool for mapping quantitative trait loci (QTL) that control developmental patterns of complex dynamic traits. Original functional mapping has been constructed within the context of simple interval mapping, without consideration of separate multiple linked QTL for a dynamic trait. In this article, we present a statistical framework for mapping QTL that affect dynamic traits by capitalizing on the strengths of functional mapping and composite interval mapping. Within this so-called composite functional-mapping framework, functional mapping models the time-dependent genetic effects of a QTL tested within a marker interval using a biologically meaningful parametric function, whereas composite interval mapping models the time-dependent genetic effects of the markers outside the test interval to control the genome background using a flexible nonparametric approach based on Legendre polynomials. Such a semiparametric framework was formulated by a maximum-likelihood model and implemented with the EM algorithm, allowing for the estimation and the test of the mathematical parameters that define the QTL effects and the regression coefficients of the Legendre polynomials that describe the marker effects. Simulation studies were performed to investigate the statistical behavior of composite functional mapping and compare its advantage in separating multiple linked QTL as compared to functional mapping. We used the new mapping approach to analyze a genetic mapping example in rice, leading to the identification of multiple QTL, some of which are linked on the same chromosome, that control the developmental trajectory of leaf age. PMID:17947431

Longer duration of homelessness is associated with a lower likelihood of non-detectable plasma HIV-1 RNA viral load among people who use illicit drugs in a Canadian setting.

PubMed

Loh, Jane; Kennedy, Mary Clare; Wood, Evan; Kerr, Thomas; Marshall, Brandon; Parashar, Surita; Montaner, Julio; Milloy, M-J

2016-11-01

Homelessness is common among people who use drugs (PWUD) and, for those living with HIV/AIDS, an important contributor to sub-optimal HIV treatment outcomes. This study aims to investigate the relationship between the duration of homelessness and the likelihood of plasma HIV-1 RNA viral load (VL) non-detectability among a cohort of HIV-positive PWUD. We used data from the ACCESS study, a long-running prospective cohort study of HIV-positive PWUD linked to comprehensive HIV clinical records including systematic plasma HIV-1 RNA VL monitoring. We estimated the longitudinal relationship between the duration of homelessness and the likelihood of exhibiting a non-detectable VL (i.e., <500 copies/mL plasma) using generalized linear mixed-effects modelling. Between May 1996 and June 2014, 922 highly active antiretroviral therapy-exposed participants were recruited and contributed 8188 observations. Of these, 4800 (59%) were characterized by non-detectable VL. Participants reported they were homeless in 910 (11%) interviews (median: six months, interquartile range: 6-12 months). A longer duration of homelessness was associated with lower odds of VL non-detectability (adjusted odds ratio = 0.71 per six-month period of homelessness, 95% confidence interval: 0.60-0.83) after adjustment for age, ancestry, drug use patterns, engagement in addiction treatment, and other potential confounders. Longer durations of episodes of homelessness in this cohort of HIV-positive illicit drug users were associated with a lower likelihood of plasma VL non-detectability. Our findings suggest that interventions that seek to promptly house homeless individuals, such as Housing First approaches, might assist in maximizing the clinical and public health benefits of antiretroviral therapy among people living with HIV/AIDS.

Dietary Acid, Age, and Serum Bicarbonate Levels among Adults in the United States

PubMed Central

Amodu, Afolarin

2013-01-01

Summary Background and objectives Greater dietary acid has been associated with lower serum bicarbonate levels in patients with CKD. Whether this association extends to the general population and if it is modified by age are unknown. Design, setting, participants, & measurements This study examined the association of the dietary acid load, estimated by net endogenous acid production, with serum bicarbonate levels in adult participants in the National Health and Nutrition Examination Survey 1999–2004. Results The mean serum bicarbonate was 24.9 mEq/L (SEM=0.1), and the mean estimated net endogenous acid production was 57.4 mEq/d (SEM=0.4). Serum bicarbonate was linearly associated with age, such that the oldest participants had the highest serum bicarbonate levels. After multivariable adjustment, participants in the highest quartile of net endogenous acid production had 0.40 mEq/L (95% confidence interval, −0.55 to −0.26) lower serum bicarbonate and a 33% (95% confidence interval, 3 to 72) higher likelihood of acidosis compared with those participants in the lowest quartile. There was a significant interaction by age of the association of net endogenous acid production with serum bicarbonate (P=0.005). Among participants 20–39, 40–59, and ≥60 years old, those participants in the highest net endogenous acid production quartile had 0.26 (95% confidence interval, −0.49 to −0.03), 0.60 (95% confidence interval, −0.92 to −0.29), and 0.49 (95% confidence interval, −0.84 to −0.14) mEq/L lower serum bicarbonate, respectively, compared with participants in the lowest quartile. Conclusion Greater dietary acid is associated with lower serum bicarbonate in the general US population, and the magnitude of this association is greater among middle-aged and elderly persons than younger adults. PMID:24052219

Mixture Rasch Models with Joint Maximum Likelihood Estimation

ERIC Educational Resources Information Center

Willse, John T.

2011-01-01

This research provides a demonstration of the utility of mixture Rasch models. Specifically, a model capable of estimating a mixture partial credit model using joint maximum likelihood is presented. Like the partial credit model, the mixture partial credit model has the beneficial feature of being appropriate for analysis of assessment data…

Bayesian Monte Carlo and Maximum Likelihood Approach for Uncertainty Estimation and Risk Management: Application to Lake Oxygen Recovery Model

EPA Science Inventory

Model uncertainty estimation and risk assessment is essential to environmental management and informed decision making on pollution mitigation strategies. In this study, we apply a probabilistic methodology, which combines Bayesian Monte Carlo simulation and Maximum Likelihood e...

The Effects of Model Misspecification and Sample Size on LISREL Maximum Likelihood Estimates.

ERIC Educational Resources Information Center

Baldwin, Beatrice

The robustness of LISREL computer program maximum likelihood estimates under specific conditions of model misspecification and sample size was examined. The population model used in this study contains one exogenous variable; three endogenous variables; and eight indicator variables, two for each latent variable. Conditions of model…

Maximum likelihood estimates, from censored data, for mixed-Weibull distributions

NASA Astrophysics Data System (ADS)

Jiang, Siyuan; Kececioglu, Dimitri

1992-06-01

A new algorithm for estimating the parameters of mixed-Weibull distributions from censored data is presented. The algorithm follows the principle of maximum likelihood estimate (MLE) through the expectation and maximization (EM) algorithm, and it is derived for both postmortem and nonpostmortem time-to-failure data. It is concluded that the concept of the EM algorithm is easy to understand and apply (only elementary statistics and calculus are required). The log-likelihood function cannot decrease after an EM sequence; this important feature was observed in all of the numerical calculations. The MLEs of the nonpostmortem data were obtained successfully for mixed-Weibull distributions with up to 14 parameters in a 5-subpopulation, mixed-Weibull distribution. Numerical examples indicate that some of the log-likelihood functions of the mixed-Weibull distributions have multiple local maxima; therefore, the algorithm should start at several initial guesses of the parameter set.

Maximum likelihood estimation and EM algorithm of Copas-like selection model for publication bias correction.

PubMed

Ning, Jing; Chen, Yong; Piao, Jin

2017-07-01

Publication bias occurs when the published research results are systematically unrepresentative of the population of studies that have been conducted, and is a potential threat to meaningful meta-analysis. The Copas selection model provides a flexible framework for correcting estimates and offers considerable insight into the publication bias. However, maximizing the observed likelihood under the Copas selection model is challenging because the observed data contain very little information on the latent variable. In this article, we study a Copas-like selection model and propose an expectation-maximization (EM) algorithm for estimation based on the full likelihood. Empirical simulation studies show that the EM algorithm and its associated inferential procedure performs well and avoids the non-convergence problem when maximizing the observed likelihood. © The Author 2017. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

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.

Quantification of variability and uncertainty for air toxic emission inventories with censored emission factor data.

PubMed

Frey, H Christopher; Zhao, Yuchao

2004-11-15

Probabilistic emission inventories were developed for urban air toxic emissions of benzene, formaldehyde, chromium, and arsenic for the example of Houston. Variability and uncertainty in emission factors were quantified for 71-97% of total emissions, depending upon the pollutant and data availability. Parametric distributions for interunit variability were fit using maximum likelihood estimation (MLE), and uncertainty in mean emission factors was estimated using parametric bootstrap simulation. For data sets containing one or more nondetected values, empirical bootstrap simulation was used to randomly sample detection limits for nondetected values and observations for sample values, and parametric distributions for variability were fit using MLE estimators for censored data. The goodness-of-fit for censored data was evaluated by comparison of cumulative distributions of bootstrap confidence intervals and empirical data. The emission inventory 95% uncertainty ranges are as small as -25% to +42% for chromium to as large as -75% to +224% for arsenic with correlated surrogates. Uncertainty was dominated by only a few source categories. Recommendations are made for future improvements to the analysis.

A model-based correction for outcome reporting bias in meta-analysis.

PubMed

Copas, John; Dwan, Kerry; Kirkham, Jamie; Williamson, Paula

2014-04-01

It is often suspected (or known) that outcomes published in medical trials are selectively reported. A systematic review for a particular outcome of interest can only include studies where that outcome was reported and so may omit, for example, a study that has considered several outcome measures but only reports those giving significant results. Using the methodology of the Outcome Reporting Bias (ORB) in Trials study of (Kirkham and others, 2010. The impact of outcome reporting bias in randomised controlled trials on a cohort of systematic reviews. British Medical Journal 340, c365), we suggest a likelihood-based model for estimating the effect of ORB on confidence intervals and p-values in meta-analysis. Correcting for bias has the effect of moving estimated treatment effects toward the null and hence more cautious assessments of significance. The bias can be very substantial, sometimes sufficient to completely overturn previous claims of significance. We re-analyze two contrasting examples, and derive a simple fixed effects approximation that can be used to give an initial estimate of the effect of ORB in practice.

Value of contrast-enhanced ultrasound in differential diagnosis of solid lesions of pancreas (SLP): A systematic review and a meta-analysis.

PubMed

Ran, Li; Zhao, Wenli; Zhao, Ye; Bu, Huaien

2017-07-01

Contrast-enhanced ultrasound (CEUS) is considered a novel method for diagnosing pancreatic cancer, but currently, there is no conclusive evidence of its accuracy. Using CEUS in discriminating between pancreatic carcinoma and other pancreatic lesions, we aimed to evaluate the diagnostic accuracy of CEUS in predicting pancreatic tumours. Relevant studies were selected from the PubMed, Cochrane Library, Elsevier, CNKI, VIP, and WANFANG databases dating from January 2006 to May 2017. The following terms were used as keywords: "pancreatic cancer" OR "pancreatic carcinoma," "contrast-enhanced ultrasonography" OR "contrast-enhanced ultrasound" OR "CEUS," and "diagnosis." The selection criteria are as follows: pancreatic carcinomas diagnosed by CEUS while the main reference standard was surgical pathology or biopsy (if it involved a clinical diagnosis, particular criteria emphasized); SonoVue or Levovist was the contrast agent; true positive, false positive, false negative, and true negative rates were obtained or calculated to construct the 2 × 2 contingency table; English or Chinese articles; at least 20 patients were enrolled in each group. The Quality Assessment for Studies of Diagnostic Accuracy was employed to evaluate the quality of articles. Pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, summary receiver-operating characteristic curves, and the area under curve were evaluated to estimate the overall diagnostic efficiency. Pooled sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio with 95% confidence intervals (CIs) were calculated with fixed-effect models. Eight of 184 records were eligible for a meta-analysis after independent scrutinization by 2 reviewers. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratios were 0.86 (95% CI 0.81-0.90), 0.75 (95% CI 0.68-0.82), 3.56 (95% CI 2.64-4.78), 0.19 (95% CI 0.13-0.27), and 22.260 (95% CI 8.980-55.177), respectively. The area under the SROC curve was 0.9088. CEUS has a satisfying pooled sensitivity and specificity for discriminating pancreatic cancer from other pancreatic lesions.

Nasal Airway Microbiota Profile and Severe Bronchiolitis in Infants: A Case-control Study.

PubMed

Hasegawa, Kohei; Linnemann, Rachel W; Mansbach, Jonathan M; Ajami, Nadim J; Espinola, Janice A; Petrosino, Joseph F; Piedra, Pedro A; Stevenson, Michelle D; Sullivan, Ashley F; Thompson, Amy D; Camargo, Carlos A

2017-11-01

Little is known about the relationship of airway microbiota with bronchiolitis in infants. We aimed to identify nasal airway microbiota profiles and to determine their association with the likelihood of bronchiolitis in infants. A case-control study was conducted. As a part of a multicenter prospective study, we collected nasal airway samples from 40 infants hospitalized with bronchiolitis. We concurrently enrolled 110 age-matched healthy controls. By applying 16S ribosomal RNA gene sequencing and an unbiased clustering approach to these 150 nasal samples, we identified microbiota profiles and determined the association of microbiota profiles with likelihood of bronchiolitis. Overall, the median age was 3 months and 56% were male. Unbiased clustering of airway microbiota identified 4 distinct profiles: Moraxella-dominant profile (37%), Corynebacterium/Dolosigranulum-dominant profile (27%), Staphylococcus-dominant profile (15%) and mixed profile (20%). Proportion of bronchiolitis was lowest in infants with Moraxella-dominant profile (14%) and highest in those with Staphylococcus-dominant profile (57%), corresponding to an odds ratio of 7.80 (95% confidence interval, 2.64-24.9; P < 0.001). In the multivariable model, the association between Staphylococcus-dominant profile and greater likelihood of bronchiolitis persisted (odds ratio for comparison with Moraxella-dominant profile, 5.16; 95% confidence interval, 1.26-22.9; P = 0.03). By contrast, Corynebacterium/Dolosigranulum-dominant profile group had low proportion of infants with bronchiolitis (17%); the likelihood of bronchiolitis in this group did not significantly differ from those with Moraxella-dominant profile in both unadjusted and adjusted analyses. In this case-control study, we identified 4 distinct nasal airway microbiota profiles in infants. Moraxella-dominant and Corynebacterium/Dolosigranulum-dominant profiles were associated with low likelihood of bronchiolitis, while Staphylococcus-dominant profile was associated with high likelihood of bronchiolitis.

Neural Mechanisms for Integrating Prior Knowledge and Likelihood in Value-Based Probabilistic Inference

PubMed Central

Ting, Chih-Chung; Yu, Chia-Chen; Maloney, Laurence T.

2015-01-01

In Bayesian decision theory, knowledge about the probabilities of possible outcomes is captured by a prior distribution and a likelihood function. The prior reflects past knowledge and the likelihood summarizes current sensory information. The two combined (integrated) form a posterior distribution that allows estimation of the probability of different possible outcomes. In this study, we investigated the neural mechanisms underlying Bayesian integration using a novel lottery decision task in which both prior knowledge and likelihood information about reward probability were systematically manipulated on a trial-by-trial basis. Consistent with Bayesian integration, as sample size increased, subjects tended to weigh likelihood information more compared with prior information. Using fMRI in humans, we found that the medial prefrontal cortex (mPFC) correlated with the mean of the posterior distribution, a statistic that reflects the integration of prior knowledge and likelihood of reward probability. Subsequent analysis revealed that both prior and likelihood information were represented in mPFC and that the neural representations of prior and likelihood in mPFC reflected changes in the behaviorally estimated weights assigned to these different sources of information in response to changes in the environment. Together, these results establish the role of mPFC in prior-likelihood integration and highlight its involvement in representing and integrating these distinct sources of information. PMID:25632152

Maximum likelihood estimation of label imperfections and its use in the identification of mislabeled patterns

NASA Technical Reports Server (NTRS)

Chittineni, C. B.

1979-01-01

The problem of estimating label imperfections and the use of the estimation in identifying mislabeled patterns is presented. Expressions for the maximum likelihood estimates of classification errors and a priori probabilities are derived from the classification of a set of labeled patterns. Expressions also are given for the asymptotic variances of probability of correct classification and proportions. Simple models are developed for imperfections in the labels and for classification errors and are used in the formulation of a maximum likelihood estimation scheme. Schemes are presented for the identification of mislabeled patterns in terms of threshold on the discriminant functions for both two-class and multiclass cases. Expressions are derived for the probability that the imperfect label identification scheme will result in a wrong decision and are used in computing thresholds. The results of practical applications of these techniques in the processing of remotely sensed multispectral data are presented.

Simple Penalties on Maximum-Likelihood Estimates of Genetic Parameters to Reduce Sampling Variation

PubMed Central

Meyer, Karin

2016-01-01

Multivariate estimates of genetic parameters are subject to substantial sampling variation, especially for smaller data sets and more than a few traits. A simple modification of standard, maximum-likelihood procedures for multivariate analyses to estimate genetic covariances is described, which can improve estimates by substantially reducing their sampling variances. This is achieved by maximizing the likelihood subject to a penalty. Borrowing from Bayesian principles, we propose a mild, default penalty—derived assuming a Beta distribution of scale-free functions of the covariance components to be estimated—rather than laboriously attempting to determine the stringency of penalization from the data. An extensive simulation study is presented, demonstrating that such penalties can yield very worthwhile reductions in loss, i.e., the difference from population values, for a wide range of scenarios and without distorting estimates of phenotypic covariances. Moreover, mild default penalties tend not to increase loss in difficult cases and, on average, achieve reductions in loss of similar magnitude to computationally demanding schemes to optimize the degree of penalization. Pertinent details required for the adaptation of standard algorithms to locate the maximum of the likelihood function are outlined. PMID:27317681

On the Performance of Maximum Likelihood versus Means and Variance Adjusted Weighted Least Squares Estimation in CFA

ERIC Educational Resources Information Center

Beauducel, Andre; Herzberg, Philipp Yorck

2006-01-01

This simulation study compared maximum likelihood (ML) estimation with weighted least squares means and variance adjusted (WLSMV) estimation. The study was based on confirmatory factor analyses with 1, 2, 4, and 8 factors, based on 250, 500, 750, and 1,000 cases, and on 5, 10, 20, and 40 variables with 2, 3, 4, 5, and 6 categories. There was no…

A maximum pseudo-profile likelihood estimator for the Cox model under length-biased sampling

PubMed Central

Huang, Chiung-Yu; Qin, Jing; Follmann, Dean A.

2012-01-01

This paper considers semiparametric estimation of the Cox proportional hazards model for right-censored and length-biased data arising from prevalent sampling. To exploit the special structure of length-biased sampling, we propose a maximum pseudo-profile likelihood estimator, which can handle time-dependent covariates and is consistent under covariate-dependent censoring. Simulation studies show that the proposed estimator is more efficient than its competitors. A data analysis illustrates the methods and theory. PMID:23843659

Modelling HIV/AIDS epidemics in sub-Saharan Africa using seroprevalence data from antenatal clinics.

PubMed Central

Salomon, J. A.; Murray, C. J.

2001-01-01

OBJECTIVE: To improve the methodological basis for modelling the HIV/AIDS epidemics in adults in sub-Saharan Africa, with examples from Botswana, Central African Republic, Ethiopia, and Zimbabwe. Understanding the magnitude and trajectory of the HIV/AIDS epidemic is essential for planning and evaluating control strategies. METHODS: Previous mathematical models were developed to estimate epidemic trends based on sentinel surveillance data from pregnant women. In this project, we have extended these models in order to take full advantage of the available data. We developed a maximum likelihood approach for the estimation of model parameters and used numerical simulation methods to compute uncertainty intervals around the estimates. FINDINGS: In the four countries analysed, there were an estimated half a million new adult HIV infections in 1999 (range: 260 to 960 thousand), 4.7 million prevalent infections (range: 3.0 to 6.6 million), and 370 thousand adult deaths from AIDS (range: 266 to 492 thousand). CONCLUSION: While this project addresses some of the limitations of previous modelling efforts, an important research agenda remains, including the need to clarify the relationship between sentinel data from pregnant women and the epidemiology of HIV and AIDS in the general population. PMID:11477962

Risk Estimation and Sexual Behaviour: A Longitudinal Study of 16- 21-year olds.

PubMed

Breakwell, G M; Breakwell, G M

1996-01-01

The relationships among risk estimation, impulsivity and patterns of sexual risk-taking in 16-21-year-olds are examined. A sample of 236 males and 340 females completed a postal questionnaire on three occasions at annual intervals. They reported their assessment of their own risk of HIV infection, the risk of HIV infection associated with six types of sexual activity, their likelihood of engaging in each of these activities, and whether they had participated in these activities between the first and second data collections. Impulsivity was indexed using a standard test. The data support the conclusion that strong social representations of sexual risks exist which do not markedly change during late adolescence. These risk estimates predict behavioural expectations, primarily for the riskiest behaviours, and for females (actual participation in vaginal sex); but for males, risk estimates fail to predict behaviour. Evidence here for a rational model of individual decision- making in relation to sexual risk- taking is sparse. Impulsivity was not a good predictor of expected or actual patterns of sexual behaviour, though higher impulsivity was associated with having more sexual partners and, in females, with starting to have sex younger.

Estimating site occupancy and abundance using indirect detection indices

USGS Publications Warehouse

Stanley, T.R.; Royle, J. Andrew

2005-01-01

Knowledge of factors influencing animal distribution and abundance is essential in many areas of ecological research, management, and policy-making. Because common methods for modeling and estimating abundance (e.g., capture-recapture, distance sampling) are sometimes not practical for large areas or elusive species, indices are sometimes used as surrogate measures of abundance. We present an extension of the Royle and Nichols (2003) generalization of the MacKenzie et al. (2002) site-occupancy model that incorporates length of the sampling interval into the, model for detection probability. As a result, we obtain a modeling framework that shows how useful information can be extracted from a class of index methods we call indirect detection indices (IDIs). Examples of IDIs include scent station, tracking tube, snow track, tracking plate, and hair snare surveys. Our model is maximum likelihood, and it can be used to estimate site occupancy and model factors influencing patterns of occupancy and abundance in space. Under certain circumstances, it can also be used to estimate abundance. We evaluated model properties using Monte Carlo simulations and illustrate the method with tracking tube and scent station data. We believe this model will be a useful tool for determining factors that influence animal distribution and abundance.

Incidence of Mastitis in the Neonatal Period in a Traditional Breastfeeding Society: Results of a Cohort Study.

PubMed

Khanal, Vishnu; Scott, Jane A; Lee, Andy H; Binns, Colin W

2015-12-01

Mastitis is a painful problem experienced by breastfeeding women, especially in the first few weeks postpartum. There have been limited studies of the incidence of mastitis from traditionally breastfeeding societies in South Asia. This study investigated the incidence, determinants, and management of mastitis in the first month postpartum, as well as its association with breastfeeding outcomes at 4 and 6 months postpartum, in western Nepal. Subjects were a subsample of 338 mothers participating in a larger prospective cohort study conducted in 2014 in western Nepal. Mothers were interviewed during the first month postpartum and again at 4 and 6 months to obtain information on breastfeeding practices. The association of mastitis and determinant variables was investigated using multivariable logistic regression, and the association with breastfeeding duration was examined using Kaplan-Meier estimation. The incidence of mastitis was 8.0% (95% confidence interval, 5.1%, 10.8%) in the first month postpartum. Prelacteal feeding (adjusted odds ratio = 2.76; 95% confidence interval, 1.03, 7.40) and cesarean section (adjusted odds ratio = 3.52; 95% confidence interval, 1.09, 11.42) were associated with a higher likelihood of mastitis. Kaplan-Meier estimation showed no significant difference in the duration of exclusive breastfeeding among the mothers who experienced an episode of mastitis and those who did not. Roughly one in 10 (8.0%) women experienced mastitis in the first month postpartum, and there appeared to be little effect of mastitis on breastfeeding outcomes. Traditional breastfeeding practices should be encouraged, and the management of mastitis should be included as a part of lactation promotion.

Quasi-Maximum Likelihood Estimation of Structural Equation Models with Multiple Interaction and Quadratic Effects

ERIC Educational Resources Information Center

Klein, Andreas G.; Muthen, Bengt O.

2007-01-01

In this article, a nonlinear structural equation model is introduced and a quasi-maximum likelihood method for simultaneous estimation and testing of multiple nonlinear effects is developed. The focus of the new methodology lies on efficiency, robustness, and computational practicability. Monte-Carlo studies indicate that the method is highly…

Estimation of Complex Generalized Linear Mixed Models for Measurement and Growth

ERIC Educational Resources Information Center

Jeon, Minjeong

2012-01-01

Maximum likelihood (ML) estimation of generalized linear mixed models (GLMMs) is technically challenging because of the intractable likelihoods that involve high dimensional integrations over random effects. The problem is magnified when the random effects have a crossed design and thus the data cannot be reduced to small independent clusters. A…

The Neural Bases of Difficult Speech Comprehension and Speech Production: Two Activation Likelihood Estimation (ALE) Meta-Analyses

ERIC Educational Resources Information Center

Adank, Patti

2012-01-01

The role of speech production mechanisms in difficult speech comprehension is the subject of on-going debate in speech science. Two Activation Likelihood Estimation (ALE) analyses were conducted on neuroimaging studies investigating difficult speech comprehension or speech production. Meta-analysis 1 included 10 studies contrasting comprehension…

Race and weight change in US women: the roles of socioeconomic and marital status.

PubMed Central

Kahn, H S; Williamson, D F; Stevens, J A

1991-01-01

BACKGROUND. The prevalence of overweight among Black women in the US is higher than among White women, but the causes are unknown. METHODS. We examined the weight change for 514 Black and 2,770 White women who entered the first Health and Nutrtion Examination Survey (1971-75) at ages 25-44 years and were weighed again a decade later. We used multivariate analyses to estimate the weight-change effectgs associated with race, family income, education, and marital change. RESULTS. After multiple adjustments, Black race, education below college level, and becoming married during the follow-up interval were each independently associated with an increased mean weight change. Using multivariate logistic analyses, Black race was not independently associated with an increased risk of major weight gain (change greater than or equal to +13 kg), but it was associated with a reduced likelihood of major weight loss (change less than or equal to -7 kg) (odds ratio - 0.64 [95% CI -0.41, 0.97])]. Very low family income was independently associated with the likelihood of both major weight gain (OR - 1.71 [95% CI - 1.15, 2.55]) and major weight loss (OR - 1.86 [95% CI - 1.18, 2.95]). CONCLUSIONS. Amont US women, Black race is independently associated with a reduced likelihood of major weight loss, but not with major weight gain. Women at greatest risk of weight gain are those with education below college level, those entering marriage, and those with very low family income. PMID:2036117

Race and weight change in US women: the roles of socioeconomic and marital status.

PubMed

Kahn, H S; Williamson, D F; Stevens, J A

1991-03-01

The prevalence of overweight among Black women in the US is higher than among White women, but the causes are unknown. We examined the weight change for 514 Black and 2,770 White women who entered the first Health and Nutrtion Examination Survey (1971-75) at ages 25-44 years and were weighed again a decade later. We used multivariate analyses to estimate the weight-change effectgs associated with race, family income, education, and marital change. After multiple adjustments, Black race, education below college level, and becoming married during the follow-up interval were each independently associated with an increased mean weight change. Using multivariate logistic analyses, Black race was not independently associated with an increased risk of major weight gain (change greater than or equal to +13 kg), but it was associated with a reduced likelihood of major weight loss (change less than or equal to -7 kg) (odds ratio - 0.64 [95% CI -0.41, 0.97])]. Very low family income was independently associated with the likelihood of both major weight gain (OR - 1.71 [95% CI - 1.15, 2.55]) and major weight loss (OR - 1.86 [95% CI - 1.18, 2.95]). Amont US women, Black race is independently associated with a reduced likelihood of major weight loss, but not with major weight gain. Women at greatest risk of weight gain are those with education below college level, those entering marriage, and those with very low family income.

Make the most of your samples: Bayes factor estimators for high-dimensional models of sequence evolution.

PubMed

Baele, Guy; Lemey, Philippe; Vansteelandt, Stijn

2013-03-06

Accurate model comparison requires extensive computation times, especially for parameter-rich models of sequence evolution. In the Bayesian framework, model selection is typically performed through the evaluation of a Bayes factor, the ratio of two marginal likelihoods (one for each model). Recently introduced techniques to estimate (log) marginal likelihoods, such as path sampling and stepping-stone sampling, offer increased accuracy over the traditional harmonic mean estimator at an increased computational cost. Most often, each model's marginal likelihood will be estimated individually, which leads the resulting Bayes factor to suffer from errors associated with each of these independent estimation processes. We here assess the original 'model-switch' path sampling approach for direct Bayes factor estimation in phylogenetics, as well as an extension that uses more samples, to construct a direct path between two competing models, thereby eliminating the need to calculate each model's marginal likelihood independently. Further, we provide a competing Bayes factor estimator using an adaptation of the recently introduced stepping-stone sampling algorithm and set out to determine appropriate settings for accurately calculating such Bayes factors, with context-dependent evolutionary models as an example. While we show that modest efforts are required to roughly identify the increase in model fit, only drastically increased computation times ensure the accuracy needed to detect more subtle details of the evolutionary process. We show that our adaptation of stepping-stone sampling for direct Bayes factor calculation outperforms the original path sampling approach as well as an extension that exploits more samples. Our proposed approach for Bayes factor estimation also has preferable statistical properties over the use of individual marginal likelihood estimates for both models under comparison. Assuming a sigmoid function to determine the path between two competing models, we provide evidence that a single well-chosen sigmoid shape value requires less computational efforts in order to approximate the true value of the (log) Bayes factor compared to the original approach. We show that the (log) Bayes factors calculated using path sampling and stepping-stone sampling differ drastically from those estimated using either of the harmonic mean estimators, supporting earlier claims that the latter systematically overestimate the performance of high-dimensional models, which we show can lead to erroneous conclusions. Based on our results, we argue that highly accurate estimation of differences in model fit for high-dimensional models requires much more computational effort than suggested in recent studies on marginal likelihood estimation.

Make the most of your samples: Bayes factor estimators for high-dimensional models of sequence evolution

PubMed Central

2013-01-01

Background Accurate model comparison requires extensive computation times, especially for parameter-rich models of sequence evolution. In the Bayesian framework, model selection is typically performed through the evaluation of a Bayes factor, the ratio of two marginal likelihoods (one for each model). Recently introduced techniques to estimate (log) marginal likelihoods, such as path sampling and stepping-stone sampling, offer increased accuracy over the traditional harmonic mean estimator at an increased computational cost. Most often, each model’s marginal likelihood will be estimated individually, which leads the resulting Bayes factor to suffer from errors associated with each of these independent estimation processes. Results We here assess the original ‘model-switch’ path sampling approach for direct Bayes factor estimation in phylogenetics, as well as an extension that uses more samples, to construct a direct path between two competing models, thereby eliminating the need to calculate each model’s marginal likelihood independently. Further, we provide a competing Bayes factor estimator using an adaptation of the recently introduced stepping-stone sampling algorithm and set out to determine appropriate settings for accurately calculating such Bayes factors, with context-dependent evolutionary models as an example. While we show that modest efforts are required to roughly identify the increase in model fit, only drastically increased computation times ensure the accuracy needed to detect more subtle details of the evolutionary process. Conclusions We show that our adaptation of stepping-stone sampling for direct Bayes factor calculation outperforms the original path sampling approach as well as an extension that exploits more samples. Our proposed approach for Bayes factor estimation also has preferable statistical properties over the use of individual marginal likelihood estimates for both models under comparison. Assuming a sigmoid function to determine the path between two competing models, we provide evidence that a single well-chosen sigmoid shape value requires less computational efforts in order to approximate the true value of the (log) Bayes factor compared to the original approach. We show that the (log) Bayes factors calculated using path sampling and stepping-stone sampling differ drastically from those estimated using either of the harmonic mean estimators, supporting earlier claims that the latter systematically overestimate the performance of high-dimensional models, which we show can lead to erroneous conclusions. Based on our results, we argue that highly accurate estimation of differences in model fit for high-dimensional models requires much more computational effort than suggested in recent studies on marginal likelihood estimation. PMID:23497171

Diagnostic value of panoramic radiography in predicting inferior alveolar nerve injury after mandibular third molar extraction: a meta-analysis.

PubMed

Liu, W; Yin, W; Zhang, R; Li, J; Zheng, Y

2015-06-01

The aim of this study was to evaluate the predictive value of panoramic radiography on inferior alveolar nerve (IAN) injury after extraction of the mandibular third molar. Relevant studies up to 1 June 2014 that discussed the association of panoramic radiography signs and post-mandibular third molar extraction IAN injury were systematically retrieved from the databases of PubMed, Embase, Springerlink, Web of Science and Cochrane library. The effect size of pooled sensitivity, specificity, positive likelihood ratios (PLR), negative likelihood ratios (NLR) and diagnostic odds ratio (DOR) with their 95% confidence intervals (CI) were statistically analysed with Meta-disc 1.4 software. Nine articles were included in this meta-analysis. The pooled estimates of sensitivity and specificity were 0.56 (95% CI: 0.50-0.61) and 0.86 (95% CI: 0.84-0.87), respectively. The overall PLR was 3.46 (95% CI: 2.02-5.92) and overall NLR was 0.58 (95% CI: 0.45-0.73). The pooled estimate of DOR was 6.49 (95% CI: 2.92-14.44). The area under the summary receiver operating characteristic curve was 0.7143 ± 0.0604. The meta-analysis indicated that interpretation of panoramic radiography based on darkening of the root had a high specificity in predicting IAN injury after mandibular third molar extraction. However, the ability of this panoramic radiography marker to detect true positive IAN injury was not satisfactory. © 2015 Australian Dental Association.

Approximated maximum likelihood estimation in multifractal random walks

NASA Astrophysics Data System (ADS)

Løvsletten, O.; Rypdal, M.

2012-04-01

We present an approximated maximum likelihood method for the multifractal random walk processes of [E. Bacry , Phys. Rev. EPLEEE81539-375510.1103/PhysRevE.64.026103 64, 026103 (2001)]. The likelihood is computed using a Laplace approximation and a truncation in the dependency structure for the latent volatility. The procedure is implemented as a package in the r computer language. Its performance is tested on synthetic data and compared to an inference approach based on the generalized method of moments. The method is applied to estimate parameters for various financial stock indices.

Residential Light and Risk for Depression and Falls: Results from the LARES Study of Eight European Cities

PubMed Central

Brown, Mary Jean; Jacobs, David E.

2011-01-01

Objectives. We examined the relationship between self-reported inadequate residential natural light and risk for depression or falls among adults aged 18 years or older. Methods. Generalized estimating equations were used to calculate the odds of depression or falls in participants with self-reported inadequate natural residential light vs. those reporting adequate light (n=6,017) using data from the World Health Organization's Large Analysis and Review of European Housing and Health Survey, a large cross-sectional study of housing and health in representative populations from eight European cities. Results. Participants reporting inadequate natural light in their dwellings were 1.4 times (95% confidence interval [CI] 1.2,1.7) as likely to report depression and 1.5 times (95% CI 1.2, 1.9) as likely to report a fall compared with those satisfied with their dwelling's light. After adjustment for major confounders, the likelihood of depression changed slightly, while the likelihood of a fall increased to 2.5 (95% CI 1.5, 4.2). Conclusion. Self-reported inadequate light in housing is independently associated with depression and falls. Increasing light in housing, a relatively inexpensive intervention, may improve two distinct health conditions. PMID:21563721

Residential light and risk for depression and falls: results from the LARES study of eight European cities.

PubMed

Brown, Mary Jean; Jacobs, David E

2011-01-01

We examined the relationship between self-reported inadequate residential natural light and risk for depression or falls among adults aged 18 years or older. Generalized estimating equations were used to calculate the odds of depression or falls in participants with self-reported inadequate natural residential light vs. those reporting adequate light (n = 6,017) using data from the World Health Organization's Large Analysis and Review of European Housing and Health Survey, a large cross-sectional study of housing and health in representative populations from eight European cities. Participants reporting inadequate natural light in their dwellings were 1.4 times (95% confidence interval [CI] 1.2,1.7) as likely to report depression and 1.5 times (95% CI 1.2, 1.9) as likely to report a fall compared with those satisfied with their dwelling's light. After adjustment for major confounders, the likelihood of depression changed slightly, while the likelihood of a fall increased to 2.5 (95% CI 1.5, 4.2). Self-reported inadequate light in housing is independently associated with depression and falls. Increasing light in housing, a relatively inexpensive intervention, may improve two distinct health conditions.

Two models for evaluating landslide hazards

USGS Publications Warehouse

Davis, J.C.; Chung, C.-J.; Ohlmacher, G.C.

2006-01-01

Two alternative procedures for estimating landslide hazards were evaluated using data on topographic digital elevation models (DEMs) and bedrock lithologies in an area adjacent to the Missouri River in Atchison County, Kansas, USA. The two procedures are based on the likelihood ratio model but utilize different assumptions. The empirical likelihood ratio model is based on non-parametric empirical univariate frequency distribution functions under an assumption of conditional independence while the multivariate logistic discriminant model assumes that likelihood ratios can be expressed in terms of logistic functions. The relative hazards of occurrence of landslides were estimated by an empirical likelihood ratio model and by multivariate logistic discriminant analysis. Predictor variables consisted of grids containing topographic elevations, slope angles, and slope aspects calculated from a 30-m DEM. An integer grid of coded bedrock lithologies taken from digitized geologic maps was also used as a predictor variable. Both statistical models yield relative estimates in the form of the proportion of total map area predicted to already contain or to be the site of future landslides. The stabilities of estimates were checked by cross-validation of results from random subsamples, using each of the two procedures. Cell-by-cell comparisons of hazard maps made by the two models show that the two sets of estimates are virtually identical. This suggests that the empirical likelihood ratio and the logistic discriminant analysis models are robust with respect to the conditional independent assumption and the logistic function assumption, respectively, and that either model can be used successfully to evaluate landslide hazards. ?? 2006.

Assessment of Likelihood Ratio for Four Contact Dermatitis Symptoms of Vinca Minor.

PubMed

Bagchi, Suman; Nayak, Chaturbhuja; Rutten, Lex; Koley, Munmun; Saha, Subhranil; Roy, Mousumi; Abbas, Sk Jahir; Ghosh, Simantini; Halder, Suman; Wani, Ghulam Nabi; Mir, Anaitulah Ahmad; Bhat, Showket Ahmad

2018-05-21

 Contact dermatitis (CD) is a frequently occurring medical condition, for which Vinca minor (VM) is one of the recommended homeopathic medicines. However, the symptoms indicating this medicine have not yet been assessed systematically. Likelihood ratio (LR), based on Bayesian statistics, may yield better estimation of a medicine's indication than the existing method of entry of symptoms into materia medica and repertories.  We investigated LRs of four CD symptoms of VM: (1) great sensitiveness of skin, with redness and soreness from slightest rubbing; (2) weeping eczema with foul, thick crusts; (3) itching amelioration in open air; and (4) CD of scalp. An observational, prospective, patient-outcome study was conducted in five different practice settings on 390 CD patients over 18 months using three outcomes-Glasgow Homeopathic Hospital Outcome Scale (GHHOS), Scoring Atopic Dermatitis (SCORAD), and Dermatology Life Quality Index (DLQI), assessed at baseline, after 3 and 6 months. The LR of each of the four symptoms was estimated as per the patient-rated outcomes on GHHOS.  Seventy-four VM and 316 non-VM cases were analyzed. Estimated LRs were as follows: symptom 1, 1.29 (95% confidence interval [CI]: 0.65 to 2.60); symptom 2, 1.48 (95% CI: 0.80 to 2.74); symptom 3, 1.70 (95% CI: 0.94 to 3.07); symptom 4, 1.36 (95% CI: 0.74 to 2.51). There were statistically significant reductions in SCORAD and DLQI scores over 3 and 6 months.  There was insufficient evidence to attribute any of the four assessed symptoms clearly to VM. Though non-significant, a high LR was observed for "itching amelioration in open air" (symptom 3). Symptoms in the homeopathic materia medica for VM are perhaps over-represented. More research of this nature is warranted. The Faculty of Homeopathy.

Climate reconstruction analysis using coexistence likelihood estimation (CRACLE): a method for the estimation of climate using vegetation.

PubMed

Harbert, Robert S; Nixon, Kevin C

2015-08-01

• Plant distributions have long been understood to be correlated with the environmental conditions to which species are adapted. Climate is one of the major components driving species distributions. Therefore, it is expected that the plants coexisting in a community are reflective of the local environment, particularly climate.• Presented here is a method for the estimation of climate from local plant species coexistence data. The method, Climate Reconstruction Analysis using Coexistence Likelihood Estimation (CRACLE), is a likelihood-based method that employs specimen collection data at a global scale for the inference of species climate tolerance. CRACLE calculates the maximum joint likelihood of coexistence given individual species climate tolerance characterization to estimate the expected climate.• Plant distribution data for more than 4000 species were used to show that this method accurately infers expected climate profiles for 165 sites with diverse climatic conditions. Estimates differ from the WorldClim global climate model by less than 1.5°C on average for mean annual temperature and less than ∼250 mm for mean annual precipitation. This is a significant improvement upon other plant-based climate-proxy methods.• CRACLE validates long hypothesized interactions between climate and local associations of plant species. Furthermore, CRACLE successfully estimates climate that is consistent with the widely used WorldClim model and therefore may be applied to the quantitative estimation of paleoclimate in future studies. © 2015 Botanical Society of America, Inc.

Evidence of seasonal variation in longitudinal growth of height in a sample of boys from Stuttgart Carlsschule, 1771-1793, using combined principal component analysis and maximum likelihood principle.

PubMed

Lehmann, A; Scheffler, Ch; Hermanussen, M

2010-02-01

Recent progress in modelling individual growth has been achieved by combining the principal component analysis and the maximum likelihood principle. This combination models growth even in incomplete sets of data and in data obtained at irregular intervals. We re-analysed late 18th century longitudinal growth of German boys from the boarding school Carlsschule in Stuttgart. The boys, aged 6-23 years, were measured at irregular 3-12 monthly intervals during the period 1771-1793. At the age of 18 years, mean height was 1652 mm, but height variation was large. The shortest boy reached 1474 mm, the tallest 1826 mm. Measured height closely paralleled modelled height, with mean difference of 4 mm, SD 7 mm. Seasonal height variation was found. Low growth rates occurred in spring and high growth rates in summer and autumn. The present study demonstrates that combining the principal component analysis and the maximum likelihood principle enables growth modelling in historic height data also. Copyright (c) 2009 Elsevier GmbH. All rights reserved.

Maximum-likelihood methods in wavefront sensing: stochastic models and likelihood functions

PubMed Central

Barrett, Harrison H.; Dainty, Christopher; Lara, David

2008-01-01

Maximum-likelihood (ML) estimation in wavefront sensing requires careful attention to all noise sources and all factors that influence the sensor data. We present detailed probability density functions for the output of the image detector in a wavefront sensor, conditional not only on wavefront parameters but also on various nuisance parameters. Practical ways of dealing with nuisance parameters are described, and final expressions for likelihoods and Fisher information matrices are derived. The theory is illustrated by discussing Shack–Hartmann sensors, and computational requirements are discussed. Simulation results show that ML estimation can significantly increase the dynamic range of a Shack–Hartmann sensor with four detectors and that it can reduce the residual wavefront error when compared with traditional methods. PMID:17206255

A population-based analysis of use and outcomes of laparoscopic bariatric surgery across socioeconomic groups in Taiwan.

PubMed

Huang, Chun-Che; Huang, Yu-Tung; Chiu, Chong-Chi

2015-11-11

With the growing development of minimally invasive techniques for the treatment of morbid obesity, laparoscopic bariatric surgery (LBS) is increasingly performed. This study aimed to assess the association between patients' socioeconomic status (SES) and the likelihood of undergoing LBS and related outcomes in Taiwan. This nationwide population-based study was conducted by using data from Taiwan's National Health Insurance Research Database. A total of 3678 morbidly obese patients aged 18 years and older who underwent conventional open bariatric surgery or LBS were identified between 2004 and 2011. Regression analyses were performed using generalized estimating equation (GEE) models to account for the nesting of patients within physician to assess patients' SES category associated with the use of LBS and related outcomes. Odds ratios (ORs) and 95 % confidence intervals (CIs) were estimated. Compared with those with medium and low SES (84.6 % and 80.2 %), patients with high SES (88.1 %) had the highest percentage of undergoing LBS (P < 0.001). After adjusting for patient demographics, institution and surgeon characteristics, the multivariate GEE analysis revealed that the highest likelihood of undergoing LBS was noted in morbidly obese patients with high SES (OR = 1.45, 95 % CI 1.10-1.90), followed by those with medium SES (OR = 1.27, 95 % CI 1.04-1.56). In addition, patients with high SES had slightly lower length of hospital stay (LOS; OR = 0.90, 95 % CI 0.82-0.99) and hospital treatment cost (OR = 0.93, 95 % CI 0.87-0.99) than their counterparts after adjustment. The increased likelihood of undergoing LBS and lower LOS and hospital treatment cost were noted among morbidly obese patients with higher SES. This finding suggests there is the need to improve clinical practice and reduce health disparities in the surgical treatment of morbidly obese patients.

Methadone Treatment for HIV Prevention—Feasibility, Retention, and Predictors of Attrition in Dar es Salaam, Tanzania: A Retrospective Cohort Study

PubMed Central

Lambdin, Barrot H.; Masao, Frank; Chang, Olivia; Kaduri, Pamela; Mbwambo, Jessie; Magimba, Ayoub; Sabuni, Norman; Bruce, R. Douglas

2014-01-01

Background. People who inject drugs (PWID) in Dar es Salaam, Tanzania, have an estimated human immunodeficiency virus (HIV) prevalence of 42%–50% compared with 6.9% among the general population. Extensive evidence supports methadone maintenance to lower morbidity, mortality, and transmission of HIV and other infectious diseases among PWID. In 2011, the Tanzanian government launched the first publicly funded methadone clinic on the mainland of sub-Saharan Africa at Muhimbili National Hospital. Methods. We conducted a retrospective cohort study of methadone-naive patients enrolling into methadone maintenance treatment. Kaplan-Meier survival curves were constructed to assess retention probability. Proportional hazards regression models were used to evaluate the association of characteristics with attrition from the methadone program. Results. Overall, 629 PWID enrolled into methadone treatment during the study. At 12 months, the proportion of clients retained in care was 57% (95% confidence interval [CI], 53%–62%). Compared with those receiving a low dose (<40 mg), clients receiving a medium (40–85 mg) (adjusted hazard ratio [aHR], 0.50 [95% CI, .37–.68]) and high (>85 mg) (aHR, 0.41 [95% CI, .29–.59]) dose of methadone had a lower likelihood of attrition, adjusting for other characteristics. Older clients (aHR, 0.53 per 10 years [95% CI, .42–.69]) and female clients (aHR, 0.50 [95% CI, .28–.90]) had a significantly lower likelihood of attrition, whereas clients who reported a history of sexual abuse (aHR, 2.84 [95% CI, 1.24–6.51]) had a significantly higher likelihood of attrition. Conclusions. Patient retention in methadone maintenance is comparable to estimates from programs in North America, Europe, and Asia. Future implementation strategies should focus on higher doses and flexible dosing strategies to optimize program retention and strengthened efforts for clients at higher risk of attrition. PMID:24855149

Model selection for identifying power-law scaling.

PubMed

Ton, Robert; Daffertshofer, Andreas

2016-08-01

Long-range temporal and spatial correlations have been reported in a remarkable number of studies. In particular power-law scaling in neural activity raised considerable interest. We here provide a straightforward algorithm not only to quantify power-law scaling but to test it against alternatives using (Bayesian) model comparison. Our algorithm builds on the well-established detrended fluctuation analysis (DFA). After removing trends of a signal, we determine its mean squared fluctuations in consecutive intervals. In contrast to DFA we use the values per interval to approximate the distribution of these mean squared fluctuations. This allows for estimating the corresponding log-likelihood as a function of interval size without presuming the fluctuations to be normally distributed, as is the case in conventional DFA. We demonstrate the validity and robustness of our algorithm using a variety of simulated signals, ranging from scale-free fluctuations with known Hurst exponents, via more conventional dynamical systems resembling exponentially correlated fluctuations, to a toy model of neural mass activity. We also illustrate its use for encephalographic signals. We further discuss confounding factors like the finite signal size. Our model comparison provides a proper means to identify power-law scaling including the range over which it is present. Copyright © 2016 Elsevier Inc. All rights reserved.

Genetic parameters for test day milk yields of first lactation Holstein cows by random regression models.

PubMed

de Melo, C M R; Packer, I U; Costa, C N; Machado, P F

2007-03-01

Covariance components for test day milk yield using 263 390 first lactation records of 32 448 Holstein cows were estimated using random regression animal models by restricted maximum likelihood. Three functions were used to adjust the lactation curve: the five-parameter logarithmic Ali and Schaeffer function (AS), the three-parameter exponential Wilmink function in its standard form (W) and in a modified form (W*), by reducing the range of covariate, and the combination of Legendre polynomial and W (LEG+W). Heterogeneous residual variance (RV) for different classes (4 and 29) of days in milk was considered in adjusting the functions. Estimates of RV were quite similar, rating from 4.15 to 5.29 kg2. Heritability estimates for AS (0.29 to 0.42), LEG+W (0.28 to 0.42) and W* (0.33 to 0.40) were similar, but heritability estimates used W (0.25 to 0.65) were highest than those estimated by the other functions, particularly at the end of lactation. Genetic correlations between milk yield on consecutive test days were close to unity, but decreased as the interval between test days increased. The AS function with homogeneous RV model had the best fit among those evaluated.

Impact of Violation of the Missing-at-Random Assumption on Full-Information Maximum Likelihood Method in Multidimensional Adaptive Testing

ERIC Educational Resources Information Center

Han, Kyung T.; Guo, Fanmin

2014-01-01

The full-information maximum likelihood (FIML) method makes it possible to estimate and analyze structural equation models (SEM) even when data are partially missing, enabling incomplete data to contribute to model estimation. The cornerstone of FIML is the missing-at-random (MAR) assumption. In (unidimensional) computerized adaptive testing…

Constrained Maximum Likelihood Estimation for Two-Level Mean and Covariance Structure Models

ERIC Educational Resources Information Center

Bentler, Peter M.; Liang, Jiajuan; Tang, Man-Lai; Yuan, Ke-Hai

2011-01-01

Maximum likelihood is commonly used for the estimation of model parameters in the analysis of two-level structural equation models. Constraints on model parameters could be encountered in some situations such as equal factor loadings for different factors. Linear constraints are the most common ones and they are relatively easy to handle in…

Computing Maximum Likelihood Estimates of Loglinear Models from Marginal Sums with Special Attention to Loglinear Item Response Theory.

ERIC Educational Resources Information Center

Kelderman, Henk

1992-01-01

Describes algorithms used in the computer program LOGIMO for obtaining maximum likelihood estimates of the parameters in loglinear models. These algorithms are also useful for the analysis of loglinear item-response theory models. Presents modified versions of the iterative proportional fitting and Newton-Raphson algorithms. Simulated data…

Recovery of Graded Response Model Parameters: A Comparison of Marginal Maximum Likelihood and Markov Chain Monte Carlo Estimation

ERIC Educational Resources Information Center

Kieftenbeld, Vincent; Natesan, Prathiba

2012-01-01

Markov chain Monte Carlo (MCMC) methods enable a fully Bayesian approach to parameter estimation of item response models. In this simulation study, the authors compared the recovery of graded response model parameters using marginal maximum likelihood (MML) and Gibbs sampling (MCMC) under various latent trait distributions, test lengths, and…

A fast ellipse extended target PHD filter using box-particle implementation

NASA Astrophysics Data System (ADS)

Zhang, Yongquan; Ji, Hongbing; Hu, Qi

2018-01-01

This paper presents a box-particle implementation of the ellipse extended target probability hypothesis density (ET-PHD) filter, called the ellipse extended target box particle PHD (EET-BP-PHD) filter, where the extended targets are described as a Poisson model developed by Gilholm et al. and the term "box" is here equivalent to the term "interval" used in interval analysis. The proposed EET-BP-PHD filter is capable of dynamically tracking multiple ellipse extended targets and estimating the target states and the number of targets, in the presence of clutter measurements, false alarms and missed detections. To derive the PHD recursion of the EET-BP-PHD filter, a suitable measurement likelihood is defined for a given partitioning cell, and the main implementation steps are presented along with the necessary box approximations and manipulations. The limitations and capabilities of the proposed EET-BP-PHD filter are illustrated by simulation examples. The simulation results show that a box-particle implementation of the ET-PHD filter can avoid the high number of particles and reduce computational burden, compared to a particle implementation of that for extended target tracking.

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.

Additive hazards regression and partial likelihood estimation for ecological monitoring data across space.

PubMed

Lin, Feng-Chang; Zhu, Jun

2012-01-01

We develop continuous-time models for the analysis of environmental or ecological monitoring data such that subjects are observed at multiple monitoring time points across space. Of particular interest are additive hazards regression models where the baseline hazard function can take on flexible forms. We consider time-varying covariates and take into account spatial dependence via autoregression in space and time. We develop statistical inference for the regression coefficients via partial likelihood. Asymptotic properties, including consistency and asymptotic normality, are established for parameter estimates under suitable regularity conditions. Feasible algorithms utilizing existing statistical software packages are developed for computation. We also consider a simpler additive hazards model with homogeneous baseline hazard and develop hypothesis testing for homogeneity. A simulation study demonstrates that the statistical inference using partial likelihood has sound finite-sample properties and offers a viable alternative to maximum likelihood estimation. For illustration, we analyze data from an ecological study that monitors bark beetle colonization of red pines in a plantation of Wisconsin.

Bayesian model selection: Evidence estimation based on DREAM simulation and bridge sampling

NASA Astrophysics Data System (ADS)

Volpi, Elena; Schoups, Gerrit; Firmani, Giovanni; Vrugt, Jasper A.

2017-04-01

Bayesian inference has found widespread application in Earth and Environmental Systems Modeling, providing an effective tool for prediction, data assimilation, parameter estimation, uncertainty analysis and hypothesis testing. Under multiple competing hypotheses, the Bayesian approach also provides an attractive alternative to traditional information criteria (e.g. AIC, BIC) for model selection. The key variable for Bayesian model selection is the evidence (or marginal likelihood) that is the normalizing constant in the denominator of Bayes theorem; while it is fundamental for model selection, the evidence is not required for Bayesian inference. It is computed for each hypothesis (model) by averaging the likelihood function over the prior parameter distribution, rather than maximizing it as by information criteria; the larger a model evidence the more support it receives among a collection of hypothesis as the simulated values assign relatively high probability density to the observed data. Hence, the evidence naturally acts as an Occam's razor, preferring simpler and more constrained models against the selection of over-fitted ones by information criteria that incorporate only the likelihood maximum. Since it is not particularly easy to estimate the evidence in practice, Bayesian model selection via the marginal likelihood has not yet found mainstream use. We illustrate here the properties of a new estimator of the Bayesian model evidence, which provides robust and unbiased estimates of the marginal likelihood; the method is coined Gaussian Mixture Importance Sampling (GMIS). GMIS uses multidimensional numerical integration of the posterior parameter distribution via bridge sampling (a generalization of importance sampling) of a mixture distribution fitted to samples of the posterior distribution derived from the DREAM algorithm (Vrugt et al., 2008; 2009). Some illustrative examples are presented to show the robustness and superiority of the GMIS estimator with respect to other commonly used approaches in the literature.

Maximum Likelihood Shift Estimation Using High Resolution Polarimetric SAR Clutter Model

NASA Astrophysics Data System (ADS)

Harant, Olivier; Bombrun, Lionel; Vasile, Gabriel; Ferro-Famil, Laurent; Gay, Michel

2011-03-01

This paper deals with a Maximum Likelihood (ML) shift estimation method in the context of High Resolution (HR) Polarimetric SAR (PolSAR) clutter. Texture modeling is exposed and the generalized ML texture tracking method is extended to the merging of various sensors. Some results on displacement estimation on the Argentiere glacier in the Mont Blanc massif using dual-pol TerraSAR-X (TSX) and quad-pol RADARSAT-2 (RS2) sensors are finally discussed.

Bias correction of risk estimates in vaccine safety studies with rare adverse events using a self-controlled case series design.

PubMed

Zeng, Chan; Newcomer, Sophia R; Glanz, Jason M; Shoup, Jo Ann; Daley, Matthew F; Hambidge, Simon J; Xu, Stanley

2013-12-15

The self-controlled case series (SCCS) method is often used to examine the temporal association between vaccination and adverse events using only data from patients who experienced such events. Conditional Poisson regression models are used to estimate incidence rate ratios, and these models perform well with large or medium-sized case samples. However, in some vaccine safety studies, the adverse events studied are rare and the maximum likelihood estimates may be biased. Several bias correction methods have been examined in case-control studies using conditional logistic regression, but none of these methods have been evaluated in studies using the SCCS design. In this study, we used simulations to evaluate 2 bias correction approaches-the Firth penalized maximum likelihood method and Cordeiro and McCullagh's bias reduction after maximum likelihood estimation-with small sample sizes in studies using the SCCS design. The simulations showed that the bias under the SCCS design with a small number of cases can be large and is also sensitive to a short risk period. The Firth correction method provides finite and less biased estimates than the maximum likelihood method and Cordeiro and McCullagh's method. However, limitations still exist when the risk period in the SCCS design is short relative to the entire observation period.

Nonparametric probability density estimation by optimization theoretic techniques

NASA Technical Reports Server (NTRS)

Scott, D. W.

1976-01-01

Two nonparametric probability density estimators are considered. The first is the kernel estimator. The problem of choosing the kernel scaling factor based solely on a random sample is addressed. An interactive mode is discussed and an algorithm proposed to choose the scaling factor automatically. The second nonparametric probability estimate uses penalty function techniques with the maximum likelihood criterion. A discrete maximum penalized likelihood estimator is proposed and is shown to be consistent in the mean square error. A numerical implementation technique for the discrete solution is discussed and examples displayed. An extensive simulation study compares the integrated mean square error of the discrete and kernel estimators. The robustness of the discrete estimator is demonstrated graphically.

Evidence-based Diagnostics: Adult Septic Arthritis

PubMed Central

Carpenter, Christopher R.; Schuur, Jeremiah D.; Everett, Worth W.; Pines, Jesse M.

2011-01-01

Background Acutely swollen or painful joints are common complaints in the emergency department (ED). Septic arthritis in adults is a challenging diagnosis, but prompt differentiation of a bacterial etiology is crucial to minimize morbidity and mortality. Objectives The objective was to perform a systematic review describing the diagnostic characteristics of history, physical examination, and bedside laboratory tests for nongonococcal septic arthritis. A secondary objective was to quantify test and treatment thresholds using derived estimates of sensitivity and specificity, as well as best-evidence diagnostic and treatment risks and anticipated benefits from appropriate therapy. Methods Two electronic search engines (PUBMED and EMBASE) were used in conjunction with a selected bibliography and scientific abstract hand search. Inclusion criteria included adult trials of patients presenting with monoarticular complaints if they reported sufficient detail to reconstruct partial or complete 2 × 2 contingency tables for experimental diagnostic test characteristics using an acceptable criterion standard. Evidence was rated by two investigators using the Quality Assessment Tool for Diagnostic Accuracy Studies (QUADAS). When more than one similarly designed trial existed for a diagnostic test, meta-analysis was conducted using a random effects model. Interval likelihood ratios (LRs) were computed when possible. To illustrate one method to quantify theoretical points in the probability of disease whereby clinicians might cease testing altogether and either withhold treatment (test threshold) or initiate definitive therapy in lieu of further diagnostics (treatment threshold), an interactive spreadsheet was designed and sample calculations were provided based on research estimates of diagnostic accuracy, diagnostic risk, and therapeutic risk/benefits. Results The prevalence of nongonococcal septic arthritis in ED patients with a single acutely painful joint is approximately 27% (95% confidence interval [CI] = 17% to 38%). With the exception of joint surgery (positive likelihood ratio [+LR] = 6.9) or skin infection overlying a prosthetic joint (+LR = 15.0), history, physical examination, and serum tests do not significantly alter posttest probability. Serum inflammatory markers such as white blood cell (WBC) counts, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) are not useful acutely. The interval LR for synovial white blood cell (sWBC) counts of 0 × 109–25 × 109/ L was 0.33; for 25 × 109–50 × 109/L, 1.06; for 50 × 109–100 × 109/L, 3.59; and exceeding 100 × 109/L, infinity. Synovial lactate may be useful to rule in or rule out the diagnosis of septic arthritis with a +LR ranging from 2.4 to infinity, and negative likelihood ratio (−LR) ranging from 0 to 0.46. Rapid polymerase chain reaction (PCR) of synovial fluid may identify the causative organism within 3 hours. Based on 56% sensitivity and 90% specificity for sWBC counts of >50 × 109/L in conjunction with best-evidence estimates for diagnosis-related risk and treatment-related risk/benefit, the arthrocentesis test threshold is 5%, with a treatment threshold of 39%. Conclusions Recent joint surgery or cellulitis overlying a prosthetic hip or knee were the only findings on history or physical examination that significantly alter the probability of nongonococcal septic arthritis. Extreme values of sWBC (>50 × 109/L) can increase, but not decrease, the probability of septic arthritis. Future ED-based diagnostic trials are needed to evaluate the role of clinical gestalt and the efficacy of nontraditional synovial markers such as lactate. PMID:21843213

Evidence-based diagnostics: adult septic arthritis.

PubMed

Carpenter, Christopher R; Schuur, Jeremiah D; Everett, Worth W; Pines, Jesse M

2011-08-01

Acutely swollen or painful joints are common complaints in the emergency department (ED). Septic arthritis in adults is a challenging diagnosis, but prompt differentiation of a bacterial etiology is crucial to minimize morbidity and mortality. The objective was to perform a systematic review describing the diagnostic characteristics of history, physical examination, and bedside laboratory tests for nongonococcal septic arthritis. A secondary objective was to quantify test and treatment thresholds using derived estimates of sensitivity and specificity, as well as best-evidence diagnostic and treatment risks and anticipated benefits from appropriate therapy. Two electronic search engines (PUBMED and EMBASE) were used in conjunction with a selected bibliography and scientific abstract hand search. Inclusion criteria included adult trials of patients presenting with monoarticular complaints if they reported sufficient detail to reconstruct partial or complete 2 × 2 contingency tables for experimental diagnostic test characteristics using an acceptable criterion standard. Evidence was rated by two investigators using the Quality Assessment Tool for Diagnostic Accuracy Studies (QUADAS). When more than one similarly designed trial existed for a diagnostic test, meta-analysis was conducted using a random effects model. Interval likelihood ratios (LRs) were computed when possible. To illustrate one method to quantify theoretical points in the probability of disease whereby clinicians might cease testing altogether and either withhold treatment (test threshold) or initiate definitive therapy in lieu of further diagnostics (treatment threshold), an interactive spreadsheet was designed and sample calculations were provided based on research estimates of diagnostic accuracy, diagnostic risk, and therapeutic risk/benefits. The prevalence of nongonococcal septic arthritis in ED patients with a single acutely painful joint is approximately 27% (95% confidence interval [CI] = 17% to 38%). With the exception of joint surgery (positive likelihood ratio [+LR] = 6.9) or skin infection overlying a prosthetic joint (+LR = 15.0), history, physical examination, and serum tests do not significantly alter posttest probability. Serum inflammatory markers such as white blood cell (WBC) counts, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) are not useful acutely. The interval LR for synovial white blood cell (sWBC) counts of 0 × 10(9)-25 × 10(9)/L was 0.33; for 25 × 10(9)-50 × 10(9)/L, 1.06; for 50 × 10(9)-100 × 10(9)/L, 3.59; and exceeding 100 × 10(9)/L, infinity. Synovial lactate may be useful to rule in or rule out the diagnosis of septic arthritis with a +LR ranging from 2.4 to infinity, and negative likelihood ratio (-LR) ranging from 0 to 0.46. Rapid polymerase chain reaction (PCR) of synovial fluid may identify the causative organism within 3 hours. Based on 56% sensitivity and 90% specificity for sWBC counts of >50 × 10(9)/L in conjunction with best-evidence estimates for diagnosis-related risk and treatment-related risk/benefit, the arthrocentesis test threshold is 5%, with a treatment threshold of 39%. Recent joint surgery or cellulitis overlying a prosthetic hip or knee were the only findings on history or physical examination that significantly alter the probability of nongonococcal septic arthritis. Extreme values of sWBC (>50 × 10(9)/L) can increase, but not decrease, the probability of septic arthritis. Future ED-based diagnostic trials are needed to evaluate the role of clinical gestalt and the efficacy of nontraditional synovial markers such as lactate. © 2011 by the Society for Academic Emergency Medicine.

A comparison of continuous- and discrete- time three-state models for rodent tumorigenicity experiments.

PubMed Central

Lindsey, J C; Ryan, L M

1994-01-01

The three-state illness-death model provides a useful way to characterize data from a rodent tumorigenicity experiment. Most parametrizations proposed recently in the literature assume discrete time for the death process and either discrete or continuous time for the tumor onset process. We compare these approaches with a third alternative that uses a piecewise continuous model on the hazards for tumor onset and death. All three models assume proportional hazards to characterize tumor lethality and the effect of dose on tumor onset and death rate. All of the models can easily be fitted using an Expectation Maximization (EM) algorithm. The piecewise continuous model is particularly appealing in this context because the complete data likelihood corresponds to a standard piecewise exponential model with tumor presence as a time-varying covariate. It can be shown analytically that differences between the parameter estimates given by each model are explained by varying assumptions about when tumor onsets, deaths, and sacrifices occur within intervals. The mixed-time model is seen to be an extension of the grouped data proportional hazards model [Mutat. Res. 24:267-278 (1981)]. We argue that the continuous-time model is preferable to the discrete- and mixed-time models because it gives reasonable estimates with relatively few intervals while still making full use of the available information. Data from the ED01 experiment illustrate the results. PMID:8187731

A comparison of maximum likelihood and other estimators of eigenvalues from several correlated Monte Carlo samples

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

Beer, M.

1980-12-01

The maximum likelihood method for the multivariate normal distribution is applied to the case of several individual eigenvalues. Correlated Monte Carlo estimates of the eigenvalue are assumed to follow this prescription and aspects of the assumption are examined. Monte Carlo cell calculations using the SAM-CE and VIM codes for the TRX-1 and TRX-2 benchmark reactors, and SAM-CE full core results are analyzed with this method. Variance reductions of a few percent to a factor of 2 are obtained from maximum likelihood estimation as compared with the simple average and the minimum variance individual eigenvalue. The numerical results verify that themore » use of sample variances and correlation coefficients in place of the corresponding population statistics still leads to nearly minimum variance estimation for a sufficient number of histories and aggregates.« less

Cosmological parameter estimation using Particle Swarm Optimization

NASA Astrophysics Data System (ADS)

Prasad, J.; Souradeep, T.

2014-03-01

Constraining parameters of a theoretical model from observational data is an important exercise in cosmology. There are many theoretically motivated models, which demand greater number of cosmological parameters than the standard model of cosmology uses, and make the problem of parameter estimation challenging. It is a common practice to employ Bayesian formalism for parameter estimation for which, in general, likelihood surface is probed. For the standard cosmological model with six parameters, likelihood surface is quite smooth and does not have local maxima, and sampling based methods like Markov Chain Monte Carlo (MCMC) method are quite successful. However, when there are a large number of parameters or the likelihood surface is not smooth, other methods may be more effective. In this paper, we have demonstrated application of another method inspired from artificial intelligence, called Particle Swarm Optimization (PSO) for estimating cosmological parameters from Cosmic Microwave Background (CMB) data taken from the WMAP satellite.

Interval mapping of high growth (hg), a major locus that increases weight gain in mice

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

Horvat, S.; Medrano, J.F.

1995-04-01

The high growth locus (hg) causes a major increase in weight gain and body size in mice. As a first step to map-based cloning of hg, we developed a genetic map of the hg-containing region using interval mapping of 403 F{sub 2} from a C57BL/6J-hghg x CAST/EiJ cross. The maximum likelihood position of hg was at the chromosome 10 marker D10Mit41 (LOD = 24.8) in the F{sub 2} females and 1.5 cM distal to D10Mit41 (LOD = 9.56) in the F{sub 2} males with corresponding LOD 2 support intervals of 3.7 and 5.4 cM, respectively. The peak LOD scores weremore » significantly higher than the estimated empirical threshold LOD values. The localization of hg by interval mapping was supported by a test cross of F{sub 2} mice recombinant between the LOD 2 support interval and the flanking marker. The interval mapping and test-cross indicate that hg is not allelic with candidate genes Igf1 or decorin (Dcn), a gene that was mapped close to hg in this study. The hg inheritance was recessive in females, although we could not reject recessive or additive inheritance in males. Possible causes for sex differences in peak LOD scores and for the distortion of transmission ratios observed in F{sub 2} males are discussed. The genetic map of the hg region will facilitate further fine mapping and cloning of hg, and allow searches for a homologous quantitative trait locus affecting growth in humans and domestic animals. 48 refs., 3 figs., 3 tabs.« less

The Extended-Image Tracking Technique Based on the Maximum Likelihood Estimation

NASA Technical Reports Server (NTRS)

Tsou, Haiping; Yan, Tsun-Yee

2000-01-01

This paper describes an extended-image tracking technique based on the maximum likelihood estimation. The target image is assume to have a known profile covering more than one element of a focal plane detector array. It is assumed that the relative position between the imager and the target is changing with time and the received target image has each of its pixels disturbed by an independent additive white Gaussian noise. When a rotation-invariant movement between imager and target is considered, the maximum likelihood based image tracking technique described in this paper is a closed-loop structure capable of providing iterative update of the movement estimate by calculating the loop feedback signals from a weighted correlation between the currently received target image and the previously estimated reference image in the transform domain. The movement estimate is then used to direct the imager to closely follow the moving target. This image tracking technique has many potential applications, including free-space optical communications and astronomy where accurate and stabilized optical pointing is essential.

A radiographic study of the mandibular third molar root development in different ethnic groups.

PubMed

Liversidge, H M; Peariasamy, K; Folayan, M O; Adeniyi, A O; Ngom, P I; Mikami, Y; Shimada, Y; Kuroe, K; Tvete, I F; Kvaal, S I

2017-12-01

The nature of differences in the timing of tooth formation between ethnic groups is important when estimating age. To calculate age of transition of the mandibular third (M3) molar tooth stages from archived dental radiographs from sub-Saharan Africa, Malaysia, Japan and two groups from London UK (Whites and Bangladeshi). The number of radiographs was 4555 (2028 males, 2527 females) with an age range 10-25 years. The left M3 was staged into Moorrees stages. A probit model was fitted to calculate mean ages for transitions between stages for males and females and each ethnic group separately. The estimated age distributions given each M3 stage was calculated. To assess differences in timing of M3 between ethnic groups, three models were proposed: a separate model for each ethnic group, a joint model and a third model combining some aspects across groups. The best model fit was tested using Bayesian and Akaikes information criteria (BIC and AIC) and log likelihood ratio test. Differences in mean ages of M3 root stages were found between ethnic groups, however all groups showed large standard deviation values. The AIC and log likelihood ratio test indicated that a separate model for each ethnic group was best. Small differences were also noted between timing of M3 between males and females, with the exception of the Malaysian group. These findings suggests that features of a reference data set (wide age range and uniform age distribution) and a Bayesian statistical approach are more important than population specific convenience samples to estimate age of an individual using M3. Some group differences were evident in M3 timing, however, this has some impact on the confidence interval of estimated age in females and little impact in males because of the large variation in age.

A meta-analysis of acute alcohol use and the risk of suicide attempt

PubMed Central

Borges, Guilherme; Bagge, Courtney; Cherpitel, Cheryl J.; Conner, Kenneth; Orozco, Ricardo; Rossow, Ingeborg

2016-01-01

Background While there are reviews reporting on the prevalence of acute use of alcohol (AUA) prior to suicide attempts, no review has used a meta-analytic approach to estimate common odds ratios (OR) of the effect of AUA on suicide attempts. We aim to report the results of the first meta-analysis of controlled epidemiological studies on acute alcohol use and suicide attempt. Methods The English language literature on Medline, PsychInfo, Google Scholar, and public-use databases was searched for original articles and critical review reports on acute alcohol use and suicide attempt for the period ranging from 1996–2015. Studies had to report an OR estimate for this association. Common odds ratios and 95% Confidence Intervals (95%CI) from random effects in meta-analyses for any acute alcohol use and 2 levels of alcohol use on suicide attempt were calculated. Results We gathered 7 studies that provided OR estimates for the likelihood of suicide attempt by acute alcohol use, compared to those who did not drink alcohol. Studies used case-control (3) and case-crossover design (4). All studies found increased ORs for acute alcohol use on suicide attempt. Meta-analysis revealed a common OR of 6.97 (95%CI=4.77, 10.17) for any acute alcohol use. Evidence from 4 studies suggests that «low levels of acute drinking» resulted in an OR of 2.71 (95%CI=1.56, 4.71) and «high levels» had much greater odds of suicide attempt, OR of 37.18 (95%CI=17.38, 79.53). Conclusions Acute use of alcohol is associated with increased likelihood of a suicide attempt, particularly at high doses. Such data should be incorporated into estimates of the burden of disease associated with alcohol use, which are currently limited by a consideration of only alcohol’s chronic effects. Future research should focus on the mechanisms through which acute use of alcohol confers risk for attempt. PMID:27928972

An Iterative Maximum a Posteriori Estimation of Proficiency Level to Detect Multiple Local Likelihood Maxima

ERIC Educational Resources Information Center

Magis, David; Raiche, Gilles

2010-01-01

In this article the authors focus on the issue of the nonuniqueness of the maximum likelihood (ML) estimator of proficiency level in item response theory (with special attention to logistic models). The usual maximum a posteriori (MAP) method offers a good alternative within that framework; however, this article highlights some drawbacks of its…

Sample size calculation for studies with grouped survival data.

PubMed

Li, Zhiguo; Wang, Xiaofei; Wu, Yuan; Owzar, Kouros

2018-06-10

Grouped survival data arise often in studies where the disease status is assessed at regular visits to clinic. The time to the event of interest can only be determined to be between two adjacent visits or is right censored at one visit. In data analysis, replacing the survival time with the endpoint or midpoint of the grouping interval leads to biased estimators of the effect size in group comparisons. Prentice and Gloeckler developed a maximum likelihood estimator for the proportional hazards model with grouped survival data and the method has been widely applied. Previous work on sample size calculation for designing studies with grouped data is based on either the exponential distribution assumption or the approximation of variance under the alternative with variance under the null. Motivated by studies in HIV trials, cancer trials and in vitro experiments to study drug toxicity, we develop a sample size formula for studies with grouped survival endpoints that use the method of Prentice and Gloeckler for comparing two arms under the proportional hazards assumption. We do not impose any distributional assumptions, nor do we use any approximation of variance of the test statistic. The sample size formula only requires estimates of the hazard ratio and survival probabilities of the event time of interest and the censoring time at the endpoints of the grouping intervals for one of the two arms. The formula is shown to perform well in a simulation study and its application is illustrated in the three motivating examples. Copyright © 2018 John Wiley & Sons, Ltd.

The Maximum Likelihood Solution for Inclination-only Data

NASA Astrophysics Data System (ADS)

Arason, P.; Levi, S.

2006-12-01

The arithmetic means of inclination-only data are known to introduce a shallowing bias. Several methods have been proposed to estimate unbiased means of the inclination along with measures of the precision. Most of the inclination-only methods were designed to maximize the likelihood function of the marginal Fisher distribution. However, the exact analytical form of the maximum likelihood function is fairly complicated, and all these methods require various assumptions and approximations that are inappropriate for many data sets. For some steep and dispersed data sets, the estimates provided by these methods are significantly displaced from the peak of the likelihood function to systematically shallower inclinations. The problem in locating the maximum of the likelihood function is partly due to difficulties in accurately evaluating the function for all values of interest. This is because some elements of the log-likelihood function increase exponentially as precision parameters increase, leading to numerical instabilities. In this study we succeeded in analytically cancelling exponential elements from the likelihood function, and we are now able to calculate its value for any location in the parameter space and for any inclination-only data set, with full accuracy. Furtermore, we can now calculate the partial derivatives of the likelihood function with desired accuracy. Locating the maximum likelihood without the assumptions required by previous methods is now straight forward. The information to separate the mean inclination from the precision parameter will be lost for very steep and dispersed data sets. It is worth noting that the likelihood function always has a maximum value. However, for some dispersed and steep data sets with few samples, the likelihood function takes its highest value on the boundary of the parameter space, i.e. at inclinations of +/- 90 degrees, but with relatively well defined dispersion. Our simulations indicate that this occurs quite frequently for certain data sets, and relatively small perturbations in the data will drive the maxima to the boundary. We interpret this to indicate that, for such data sets, the information needed to separate the mean inclination and the precision parameter is permanently lost. To assess the reliability and accuracy of our method we generated large number of random Fisher-distributed data sets and used seven methods to estimate the mean inclination and precision paramenter. These comparisons are described by Levi and Arason at the 2006 AGU Fall meeting. The results of the various methods is very favourable to our new robust maximum likelihood method, which, on average, is the most reliable, and the mean inclination estimates are the least biased toward shallow values. Further information on our inclination-only analysis can be obtained from: http://www.vedur.is/~arason/paleomag

Occupancy Modeling Species-Environment Relationships with Non-ignorable Survey Designs.

PubMed

Irvine, Kathryn M; Rodhouse, Thomas J; Wright, Wilson J; Olsen, Anthony R

2018-05-26

Statistical models supporting inferences about species occurrence patterns in relation to environmental gradients are fundamental to ecology and conservation biology. A common implicit assumption is that the sampling design is ignorable and does not need to be formally accounted for in analyses. The analyst assumes data are representative of the desired population and statistical modeling proceeds. However, if datasets from probability and non-probability surveys are combined or unequal selection probabilities are used, the design may be non ignorable. We outline the use of pseudo-maximum likelihood estimation for site-occupancy models to account for such non-ignorable survey designs. This estimation method accounts for the survey design by properly weighting the pseudo-likelihood equation. In our empirical example, legacy and newer randomly selected locations were surveyed for bats to bridge a historic statewide effort with an ongoing nationwide program. We provide a worked example using bat acoustic detection/non-detection data and show how analysts can diagnose whether their design is ignorable. Using simulations we assessed whether our approach is viable for modeling datasets composed of sites contributed outside of a probability design Pseudo-maximum likelihood estimates differed from the usual maximum likelihood occu31 pancy estimates for some bat species. Using simulations we show the maximum likelihood estimator of species-environment relationships with non-ignorable sampling designs was biased, whereas the pseudo-likelihood estimator was design-unbiased. However, in our simulation study the designs composed of a large proportion of legacy or non-probability sites resulted in estimation issues for standard errors. These issues were likely a result of highly variable weights confounded by small sample sizes (5% or 10% sampling intensity and 4 revisits). Aggregating datasets from multiple sources logically supports larger sample sizes and potentially increases spatial extents for statistical inferences. Our results suggest that ignoring the mechanism for how locations were selected for data collection (e.g., the sampling design) could result in erroneous model-based conclusions. Therefore, in order to ensure robust and defensible recommendations for evidence-based conservation decision-making, the survey design information in addition to the data themselves must be available for analysts. Details for constructing the weights used in estimation and code for implementation are provided. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.

Collinear Latent Variables in Multilevel Confirmatory Factor Analysis

PubMed Central

van de Schoot, Rens; Hox, Joop

2014-01-01

Because variables may be correlated in the social and behavioral sciences, multicollinearity might be problematic. This study investigates the effect of collinearity manipulated in within and between levels of a two-level confirmatory factor analysis by Monte Carlo simulation. Furthermore, the influence of the size of the intraclass correlation coefficient (ICC) and estimation method; maximum likelihood estimation with robust chi-squares and standard errors and Bayesian estimation, on the convergence rate are investigated. The other variables of interest were rate of inadmissible solutions and the relative parameter and standard error bias on the between level. The results showed that inadmissible solutions were obtained when there was between level collinearity and the estimation method was maximum likelihood. In the within level multicollinearity condition, all of the solutions were admissible but the bias values were higher compared with the between level collinearity condition. Bayesian estimation appeared to be robust in obtaining admissible parameters but the relative bias was higher than for maximum likelihood estimation. Finally, as expected, high ICC produced less biased results compared to medium ICC conditions. PMID:29795827

How much to trust the senses: Likelihood learning

PubMed Central

Sato, Yoshiyuki; Kording, Konrad P.

2014-01-01

Our brain often needs to estimate unknown variables from imperfect information. Our knowledge about the statistical distributions of quantities in our environment (called priors) and currently available information from sensory inputs (called likelihood) are the basis of all Bayesian models of perception and action. While we know that priors are learned, most studies of prior-likelihood integration simply assume that subjects know about the likelihood. However, as the quality of sensory inputs change over time, we also need to learn about new likelihoods. Here, we show that human subjects readily learn the distribution of visual cues (likelihood function) in a way that can be predicted by models of statistically optimal learning. Using a likelihood that depended on color context, we found that a learned likelihood generalized to new priors. Thus, we conclude that subjects learn about likelihood. PMID:25398975

Variation in Cesarean Birth Rates by Labor and Delivery Nurses.

PubMed

Edmonds, Joyce K; O'Hara, Michele; Clarke, Sean P; Shah, Neel T

To examine variation in the cesarean birth rates of women cared for by labor and delivery nurses. Retrospective cohort study. One high-volume labor and delivery unit at an academic medical center in a major metropolitan area. Labor and delivery nurses who cared for nulliparous women who gave birth to term, singleton fetuses in vertex presentation. Data were extracted from electronic hospital birth records from January 1, 2013 through June 30, 2015. Cesarean rates for individual nurses were calculated based on the number of women they attended who gave birth by cesarean. Nurses were grouped into quartiles by their cesarean rates, and the effect of these rates on the likelihood of cesarean birth was estimated by a logit regression model adjusting for patient-level characteristics and clustering of births within nurses. Seventy-two nurses attended 3,031 births. The mean nurse cesarean rate was 26% (95% confidence interval [23.9, 28.1]) and ranged from 8.3% to 48%. The adjusted odds of cesarean for births attended by nurses in the highest quartile was nearly 3 times (odds ratio = 2.73, 95% confidence interval [2.3, 3.3]) greater than for births attended by nurses in the lowest quartile. The labor and delivery nurse assigned to a woman may influence the likelihood of cesarean birth. Nurse-level cesarean birth data could be used to design practice improvement initiatives to improve nurse performance. More precise measurement of the relative influence of nurses on mode of birth is needed. Copyright © 2017 AWHONN, the Association of Women's Health, Obstetric and Neonatal Nurses. Published by Elsevier Inc. All rights reserved.

A Penalized Likelihood Framework For High-Dimensional Phylogenetic Comparative Methods And An Application To New-World Monkeys Brain Evolution.

PubMed

Julien, Clavel; Leandro, Aristide; Hélène, Morlon

2018-06-19

Working with high-dimensional phylogenetic comparative datasets is challenging because likelihood-based multivariate methods suffer from low statistical performances as the number of traits p approaches the number of species n and because some computational complications occur when p exceeds n. Alternative phylogenetic comparative methods have recently been proposed to deal with the large p small n scenario but their use and performances are limited. Here we develop a penalized likelihood framework to deal with high-dimensional comparative datasets. We propose various penalizations and methods for selecting the intensity of the penalties. We apply this general framework to the estimation of parameters (the evolutionary trait covariance matrix and parameters of the evolutionary model) and model comparison for the high-dimensional multivariate Brownian (BM), Early-burst (EB), Ornstein-Uhlenbeck (OU) and Pagel's lambda models. We show using simulations that our penalized likelihood approach dramatically improves the estimation of evolutionary trait covariance matrices and model parameters when p approaches n, and allows for their accurate estimation when p equals or exceeds n. In addition, we show that penalized likelihood models can be efficiently compared using Generalized Information Criterion (GIC). We implement these methods, as well as the related estimation of ancestral states and the computation of phylogenetic PCA in the R package RPANDA and mvMORPH. Finally, we illustrate the utility of the new proposed framework by evaluating evolutionary models fit, analyzing integration patterns, and reconstructing evolutionary trajectories for a high-dimensional 3-D dataset of brain shape in the New World monkeys. We find a clear support for an Early-burst model suggesting an early diversification of brain morphology during the ecological radiation of the clade. Penalized likelihood offers an efficient way to deal with high-dimensional multivariate comparative data.

Each procedure matters: threshold for surgeon volume to minimize complications and decrease cost associated with adrenalectomy.

PubMed

Anderson, Kevin L; Thomas, Samantha M; Adam, Mohamed A; Pontius, Lauren N; Stang, Michael T; Scheri, Randall P; Roman, Sanziana A; Sosa, Julie A

2018-01-01

An association has been suggested between increasing surgeon volume and improved patient outcomes, but a threshold has not been defined for what constitutes a "high-volume" adrenal surgeon. Adult patients who underwent adrenalectomy by an identifiable surgeon between 1998-2009 were selected from the Healthcare Cost and Utilization Project National Inpatient Sample. Logistic regression modeling with restricted cubic splines was utilized to estimate the association between annual surgeon volume and complication rates in order to identify a volume threshold. A total of 3,496 surgeons performed adrenalectomies on 6,712 patients; median annual surgeon volume was 1 case. After adjustment, the likelihood of experiencing a complication decreased with increasing annual surgeon volume up to 5.6 cases (95% confidence interval, 3.27-5.96). After adjustment, patients undergoing resection by low-volume surgeons (<6 cases/year) were more likely to experience complications (odds ratio 1.71, 95% confidence interval, 1.27-2.31, P = .005), have a greater hospital stay (relative risk 1.46, 95% confidence interval, 1.25-1.70, P = .003), and at increased cost (+26.2%, 95% confidence interval, 12.6-39.9, P = .02). This study suggests that an annual threshold of surgeon volume (≥6 cases/year) that is associated with improved patient outcomes and decreased hospital cost. This volume threshold has implications for quality improvement, surgical referral and reimbursement, and surgical training. Copyright © 2017 Elsevier Inc. All rights reserved.

A robust design mark-resight abundance estimator allowing heterogeneity in resighting probabilities

USGS Publications Warehouse

McClintock, B.T.; White, Gary C.; Burnham, K.P.

2006-01-01

This article introduces the beta-binomial estimator (BBE), a closed-population abundance mark-resight model combining the favorable qualities of maximum likelihood theory and the allowance of individual heterogeneity in sighting probability (p). The model may be parameterized for a robust sampling design consisting of multiple primary sampling occasions where closure need not be met between primary occasions. We applied the model to brown bear data from three study areas in Alaska and compared its performance to the joint hypergeometric estimator (JHE) and Bowden's estimator (BOWE). BBE estimates suggest heterogeneity levels were non-negligible and discourage the use of JHE for these data. Compared to JHE and BOWE, confidence intervals were considerably shorter for the AICc model-averaged BBE. To evaluate the properties of BBE relative to JHE and BOWE when sample sizes are small, simulations were performed with data from three primary occasions generated under both individual heterogeneity and temporal variation in p. All models remained consistent regardless of levels of variation in p. In terms of precision, the AICc model-averaged BBE showed advantages over JHE and BOWE when heterogeneity was present and mean sighting probabilities were similar between primary occasions. Based on the conditions examined, BBE is a reliable alternative to JHE or BOWE and provides a framework for further advances in mark-resight abundance estimation. ?? 2006 American Statistical Association and the International Biometric Society.

Method and system for diagnostics of apparatus

NASA Technical Reports Server (NTRS)

Gorinevsky, Dimitry (Inventor)

2012-01-01

Proposed is a method, implemented in software, for estimating fault state of an apparatus outfitted with sensors. At each execution period the method processes sensor data from the apparatus to obtain a set of parity parameters, which are further used for estimating fault state. The estimation method formulates a convex optimization problem for each fault hypothesis and employs a convex solver to compute fault parameter estimates and fault likelihoods for each fault hypothesis. The highest likelihoods and corresponding parameter estimates are transmitted to a display device or an automated decision and control system. The obtained accurate estimate of fault state can be used to improve safety, performance, or maintenance processes for the apparatus.

dPIRPLE: a joint estimation framework for deformable registration and penalized-likelihood CT image reconstruction using prior images

NASA Astrophysics Data System (ADS)

Dang, H.; Wang, A. S.; Sussman, Marc S.; Siewerdsen, J. H.; Stayman, J. W.

2014-09-01

Sequential imaging studies are conducted in many clinical scenarios. Prior images from previous studies contain a great deal of patient-specific anatomical information and can be used in conjunction with subsequent imaging acquisitions to maintain image quality while enabling radiation dose reduction (e.g., through sparse angular sampling, reduction in fluence, etc). However, patient motion between images in such sequences results in misregistration between the prior image and current anatomy. Existing prior-image-based approaches often include only a simple rigid registration step that can be insufficient for capturing complex anatomical motion, introducing detrimental effects in subsequent image reconstruction. In this work, we propose a joint framework that estimates the 3D deformation between an unregistered prior image and the current anatomy (based on a subsequent data acquisition) and reconstructs the current anatomical image using a model-based reconstruction approach that includes regularization based on the deformed prior image. This framework is referred to as deformable prior image registration, penalized-likelihood estimation (dPIRPLE). Central to this framework is the inclusion of a 3D B-spline-based free-form-deformation model into the joint registration-reconstruction objective function. The proposed framework is solved using a maximization strategy whereby alternating updates to the registration parameters and image estimates are applied allowing for improvements in both the registration and reconstruction throughout the optimization process. Cadaver experiments were conducted on a cone-beam CT testbench emulating a lung nodule surveillance scenario. Superior reconstruction accuracy and image quality were demonstrated using the dPIRPLE algorithm as compared to more traditional reconstruction methods including filtered backprojection, penalized-likelihood estimation (PLE), prior image penalized-likelihood estimation (PIPLE) without registration, and prior image penalized-likelihood estimation with rigid registration of a prior image (PIRPLE) over a wide range of sampling sparsity and exposure levels.

Likelihood ratios for glaucoma diagnosis using spectral-domain optical coherence tomography.

PubMed

Lisboa, Renato; Mansouri, Kaweh; Zangwill, Linda M; Weinreb, Robert N; Medeiros, Felipe A

2013-11-01

To present a methodology for calculating likelihood ratios for glaucoma diagnosis for continuous retinal nerve fiber layer (RNFL) thickness measurements from spectral-domain optical coherence tomography (spectral-domain OCT). Observational cohort study. A total of 262 eyes of 187 patients with glaucoma and 190 eyes of 100 control subjects were included in the study. Subjects were recruited from the Diagnostic Innovations Glaucoma Study. Eyes with preperimetric and perimetric glaucomatous damage were included in the glaucoma group. The control group was composed of healthy eyes with normal visual fields from subjects recruited from the general population. All eyes underwent RNFL imaging with Spectralis spectral-domain OCT. Likelihood ratios for glaucoma diagnosis were estimated for specific global RNFL thickness measurements using a methodology based on estimating the tangents to the receiver operating characteristic (ROC) curve. Likelihood ratios could be determined for continuous values of average RNFL thickness. Average RNFL thickness values lower than 86 μm were associated with positive likelihood ratios (ie, likelihood ratios greater than 1), whereas RNFL thickness values higher than 86 μm were associated with negative likelihood ratios (ie, likelihood ratios smaller than 1). A modified Fagan nomogram was provided to assist calculation of posttest probability of disease from the calculated likelihood ratios and pretest probability of disease. The methodology allowed calculation of likelihood ratios for specific RNFL thickness values. By avoiding arbitrary categorization of test results, it potentially allows for an improved integration of test results into diagnostic clinical decision making. Copyright © 2013. Published by Elsevier Inc.

Multimodal Likelihoods in Educational Assessment: Will the Real Maximum Likelihood Score Please Stand up?

ERIC Educational Resources Information Center

Wothke, Werner; Burket, George; Chen, Li-Sue; Gao, Furong; Shu, Lianghua; Chia, Mike

2011-01-01

It has been known for some time that item response theory (IRT) models may exhibit a likelihood function of a respondent's ability which may have multiple modes, flat modes, or both. These conditions, often associated with guessing of multiple-choice (MC) questions, can introduce uncertainty and bias to ability estimation by maximum likelihood…

F-8C adaptive flight control extensions. [for maximum likelihood estimation

NASA Technical Reports Server (NTRS)

Stein, G.; Hartmann, G. L.

1977-01-01

An adaptive concept which combines gain-scheduled control laws with explicit maximum likelihood estimation (MLE) identification to provide the scheduling values is described. The MLE algorithm was improved by incorporating attitude data, estimating gust statistics for setting filter gains, and improving parameter tracking during changing flight conditions. A lateral MLE algorithm was designed to improve true air speed and angle of attack estimates during lateral maneuvers. Relationships between the pitch axis sensors inherent in the MLE design were examined and used for sensor failure detection. Design details and simulation performance are presented for each of the three areas investigated.

Quantum state estimation when qubits are lost: a no-data-left-behind approach

DOE PAGES

Williams, Brian P.; Lougovski, Pavel

2017-04-06

We present an approach to Bayesian mean estimation of quantum states using hyperspherical parametrization and an experiment-specific likelihood which allows utilization of all available data, even when qubits are lost. With this method, we report the first closed-form Bayesian mean and maximum likelihood estimates for the ideal single qubit. Due to computational constraints, we utilize numerical sampling to determine the Bayesian mean estimate for a photonic two-qubit experiment in which our novel analysis reduces burdens associated with experimental asymmetries and inefficiencies. This method can be applied to quantum states of any dimension and experimental complexity.

Estimation of Dynamic Discrete Choice Models by Maximum Likelihood and the Simulated Method of Moments

PubMed Central

Eisenhauer, Philipp; Heckman, James J.; Mosso, Stefano

2015-01-01

We compare the performance of maximum likelihood (ML) and simulated method of moments (SMM) estimation for dynamic discrete choice models. We construct and estimate a simplified dynamic structural model of education that captures some basic features of educational choices in the United States in the 1980s and early 1990s. We use estimates from our model to simulate a synthetic dataset and assess the ability of ML and SMM to recover the model parameters on this sample. We investigate the performance of alternative tuning parameters for SMM. PMID:26494926

Predicting the risk of malignancy in adnexal masses based on the Simple Rules from the International Ovarian Tumor Analysis group.

PubMed

Timmerman, Dirk; Van Calster, Ben; Testa, Antonia; Savelli, Luca; Fischerova, Daniela; Froyman, Wouter; Wynants, Laure; Van Holsbeke, Caroline; Epstein, Elisabeth; Franchi, Dorella; Kaijser, Jeroen; Czekierdowski, Artur; Guerriero, Stefano; Fruscio, Robert; Leone, Francesco P G; Rossi, Alberto; Landolfo, Chiara; Vergote, Ignace; Bourne, Tom; Valentin, Lil

2016-04-01

Accurate methods to preoperatively characterize adnexal tumors are pivotal for optimal patient management. A recent metaanalysis concluded that the International Ovarian Tumor Analysis algorithms such as the Simple Rules are the best approaches to preoperatively classify adnexal masses as benign or malignant. We sought to develop and validate a model to predict the risk of malignancy in adnexal masses using the ultrasound features in the Simple Rules. This was an international cross-sectional cohort study involving 22 oncology centers, referral centers for ultrasonography, and general hospitals. We included consecutive patients with an adnexal tumor who underwent a standardized transvaginal ultrasound examination and were selected for surgery. Data on 5020 patients were recorded in 3 phases from 2002 through 2012. The 5 Simple Rules features indicative of a benign tumor (B-features) and the 5 features indicative of malignancy (M-features) are based on the presence of ascites, tumor morphology, and degree of vascularity at ultrasonography. Gold standard was the histopathologic diagnosis of the adnexal mass (pathologist blinded to ultrasound findings). Logistic regression analysis was used to estimate the risk of malignancy based on the 10 ultrasound features and type of center. The diagnostic performance was evaluated by area under the receiver operating characteristic curve, sensitivity, specificity, positive likelihood ratio (LR+), negative likelihood ratio (LR-), positive predictive value (PPV), negative predictive value (NPV), and calibration curves. Data on 4848 patients were analyzed. The malignancy rate was 43% (1402/3263) in oncology centers and 17% (263/1585) in other centers. The area under the receiver operating characteristic curve on validation data was very similar in oncology centers (0.917; 95% confidence interval, 0.901-0.931) and other centers (0.916; 95% confidence interval, 0.873-0.945). Risk estimates showed good calibration. In all, 23% of patients in the validation data set had a very low estimated risk (<1%) and 48% had a high estimated risk (≥30%). For the 1% risk cutoff, sensitivity was 99.7%, specificity 33.7%, LR+ 1.5, LR- 0.010, PPV 44.8%, and NPV 98.9%. For the 30% risk cutoff, sensitivity was 89.0%, specificity 84.7%, LR+ 5.8, LR- 0.13, PPV 75.4%, and NPV 93.9%. Quantification of the risk of malignancy based on the Simple Rules has good diagnostic performance both in oncology centers and other centers. A simple classification based on these risk estimates may form the basis of a clinical management system. Patients with a high risk may benefit from surgery by a gynecological oncologist, while patients with a lower risk may be managed locally. Copyright © 2016 Elsevier Inc. All rights reserved.

Incidence of Mastitis in the Neonatal Period in a Traditional Breastfeeding Society: Results of a Cohort Study

PubMed Central

Scott, Jane A.; Lee, Andy H.; Binns, Colin W.

2015-01-01

Abstract Background: Mastitis is a painful problem experienced by breastfeeding women, especially in the first few weeks postpartum. There have been limited studies of the incidence of mastitis from traditionally breastfeeding societies in South Asia. This study investigated the incidence, determinants, and management of mastitis in the first month postpartum, as well as its association with breastfeeding outcomes at 4 and 6 months postpartum, in western Nepal. Subjects and Methods: Subjects were a subsample of 338 mothers participating in a larger prospective cohort study conducted in 2014 in western Nepal. Mothers were interviewed during the first month postpartum and again at 4 and 6 months to obtain information on breastfeeding practices. The association of mastitis and determinant variables was investigated using multivariable logistic regression, and the association with breastfeeding duration was examined using Kaplan–Meier estimation. Results: The incidence of mastitis was 8.0% (95% confidence interval, 5.1%, 10.8%) in the first month postpartum. Prelacteal feeding (adjusted odds ratio = 2.76; 95% confidence interval, 1.03, 7.40) and cesarean section (adjusted odds ratio = 3.52; 95% confidence interval, 1.09, 11.42) were associated with a higher likelihood of mastitis. Kaplan–Meier estimation showed no significant difference in the duration of exclusive breastfeeding among the mothers who experienced an episode of mastitis and those who did not. Conclusions: Roughly one in 10 (8.0%) women experienced mastitis in the first month postpartum, and there appeared to be little effect of mastitis on breastfeeding outcomes. Traditional breastfeeding practices should be encouraged, and the management of mastitis should be included as a part of lactation promotion. PMID:26488802

Estimates of over-diagnosis of breast cancer due to population-based mammography screening in South Australia after adjustment for lead time effects.

PubMed

Beckmann, Kerri; Duffy, Stephen W; Lynch, John; Hiller, Janet; Farshid, Gelareh; Roder, David

2015-09-01

To estimate over-diagnosis due to population-based mammography screening using a lead time adjustment approach, with lead time measures based on symptomatic cancers only. Women aged 40-84 in 1989-2009 in South Australia eligible for mammography screening. Numbers of observed and expected breast cancer cases were compared, after adjustment for lead time. Lead time effects were modelled using age-specific estimates of lead time (derived from interval cancer rates and predicted background incidence, using maximum likelihood methods) and screening sensitivity, projected background breast cancer incidence rates (in the absence of screening), and proportions screened, by age and calendar year. Lead time estimates were 12, 26, 43 and 53 months, for women aged 40-49, 50-59, 60-69 and 70-79 respectively. Background incidence rates were estimated to have increased by 0.9% and 1.2% per year for invasive and all breast cancer. Over-diagnosis among women aged 40-84 was estimated at 7.9% (0.1-12.0%) for invasive cases and 12.0% (5.7-15.4%) when including ductal carcinoma in-situ (DCIS). We estimated 8% over-diagnosis for invasive breast cancer and 12% inclusive of DCIS cancers due to mammography screening among women aged 40-84. These estimates may overstate the extent of over-diagnosis if the increasing prevalence of breast cancer risk factors has led to higher background incidence than projected. © The Author(s) 2015.

Empirical best linear unbiased prediction method for small areas with restricted maximum likelihood and bootstrap procedure to estimate the average of household expenditure per capita in Banjar Regency

NASA Astrophysics Data System (ADS)

Aminah, Agustin Siti; Pawitan, Gandhi; Tantular, Bertho

2017-03-01

So far, most of the data published by Statistics Indonesia (BPS) as data providers for national statistics are still limited to the district level. Less sufficient sample size for smaller area levels to make the measurement of poverty indicators with direct estimation produced high standard error. Therefore, the analysis based on it is unreliable. To solve this problem, the estimation method which can provide a better accuracy by combining survey data and other auxiliary data is required. One method often used for the estimation is the Small Area Estimation (SAE). There are many methods used in SAE, one of them is Empirical Best Linear Unbiased Prediction (EBLUP). EBLUP method of maximum likelihood (ML) procedures does not consider the loss of degrees of freedom due to estimating β with β ^. This drawback motivates the use of the restricted maximum likelihood (REML) procedure. This paper proposed EBLUP with REML procedure for estimating poverty indicators by modeling the average of household expenditures per capita and implemented bootstrap procedure to calculate MSE (Mean Square Error) to compare the accuracy EBLUP method with the direct estimation method. Results show that EBLUP method reduced MSE in small area estimation.

On the Existence and Uniqueness of JML Estimates for the Partial Credit Model

ERIC Educational Resources Information Center

Bertoli-Barsotti, Lucio

2005-01-01

A necessary and sufficient condition is given in this paper for the existence and uniqueness of the maximum likelihood (the so-called joint maximum likelihood) estimate of the parameters of the Partial Credit Model. This condition is stated in terms of a structural property of the pattern of the data matrix that can be easily verified on the basis…

Formulating the Rasch Differential Item Functioning Model under the Marginal Maximum Likelihood Estimation Context and Its Comparison with Mantel-Haenszel Procedure in Short Test and Small Sample Conditions

ERIC Educational Resources Information Center

Paek, Insu; Wilson, Mark

2011-01-01

This study elaborates the Rasch differential item functioning (DIF) model formulation under the marginal maximum likelihood estimation context. Also, the Rasch DIF model performance was examined and compared with the Mantel-Haenszel (MH) procedure in small sample and short test length conditions through simulations. The theoretically known…

Use of Bayes theorem to correct size-specific sampling bias in growth data.

PubMed

Troynikov, V S

1999-03-01

The bayesian decomposition of posterior distribution was used to develop a likelihood function to correct bias in the estimates of population parameters from data collected randomly with size-specific selectivity. Positive distributions with time as a parameter were used for parametrization of growth data. Numerical illustrations are provided. The alternative applications of the likelihood to estimate selectivity parameters are discussed.

ATAC Autocuer Modeling Analysis.

DTIC Science & Technology

1981-01-01

the analysis of the simple rectangular scrnentation (1) is based on detection and estimation theory (2). This approach uses the concept of maximum ...continuous wave forms. In order to develop the principles of maximum likelihood, it is con- venient to develop the principles for the "classical...the concept of maximum likelihood is significant in that it provides the optimum performance of the detection/estimation problem. With a knowledge of

Factors Associated With Breastfeeding Duration Among Connecticut Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) Participants

PubMed Central

Haughton, Jannett; Gregorio, David; Pérez-Escamilla, Rafael

2011-01-01

This retrospective study aimed to identify factors associated with breastfeeding duration among women enrolled in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) of Hartford, Connecticut. The authors included mothers whose children were younger than 5 years and had stopped breastfeeding (N = 155). Women who had planned their pregnancies were twice as likely as those who did not plan them to breastfeed for more than 6 months (odds ratio, 2.15; 95% confidence interval, 1.00–4.64). One additional year of maternal age was associated with a 9% increase on the likelihood of breastfeeding for more than 6 months (odds ratio, 1.09; 95% confidence interval, 1.02–1.17). Time in the United States was inversely associated with the likelihood of breastfeeding for more than 6 months (odds ratio, 0.96; 95% confidence interval, 0.92–0.99). Return to work, sore nipples, lack of access to breast pumps, and free formula provided by WIC were identified as breastfeeding barriers. Findings can help WIC improve its breastfeeding promotion efforts. PMID:20689103

Quantifying the Establishment Likelihood of Invasive Alien Species Introductions Through Ports with Application to Honeybees in Australia.

PubMed

Heersink, Daniel K; Caley, Peter; Paini, Dean R; Barry, Simon C

2016-05-01

The cost of an uncontrolled incursion of invasive alien species (IAS) arising from undetected entry through ports can be substantial, and knowledge of port-specific risks is needed to help allocate limited surveillance resources. Quantifying the establishment likelihood of such an incursion requires quantifying the ability of a species to enter, establish, and spread. Estimation of the approach rate of IAS into ports provides a measure of likelihood of entry. Data on the approach rate of IAS are typically sparse, and the combinations of risk factors relating to country of origin and port of arrival diverse. This presents challenges to making formal statistical inference on establishment likelihood. Here we demonstrate how these challenges can be overcome with judicious use of mixed-effects models when estimating the incursion likelihood into Australia of the European (Apis mellifera) and Asian (A. cerana) honeybees, along with the invasive parasites of biosecurity concern they host (e.g., Varroa destructor). Our results demonstrate how skewed the establishment likelihood is, with one-tenth of the ports accounting for 80% or more of the likelihood for both species. These results have been utilized by biosecurity agencies in the allocation of resources to the surveillance of maritime ports. © 2015 Society for Risk Analysis.

An evaluation of portion size estimation aids: precision, ease of use and likelihood of future use.

PubMed

Faulkner, Gemma P; Livingstone, M Barbara E; Pourshahidi, L Kirsty; Spence, Michelle; Dean, Moira; O'Brien, Sinead; Gibney, Eileen R; Wallace, Julie Mw; McCaffrey, Tracy A; Kerr, Maeve A

2016-09-01

The present study aimed to evaluate the precision, ease of use and likelihood of future use of portion size estimation aids (PSEA). A range of PSEA were used to estimate the serving sizes of a range of commonly eaten foods and rated for ease of use and likelihood of future usage. For each food, participants selected their preferred PSEA from a range of options including: quantities and measures; reference objects; measuring; and indicators on food packets. These PSEA were used to serve out various foods (e.g. liquid, amorphous, and composite dishes). Ease of use and likelihood of future use were noted. The foods were weighed to determine the precision of each PSEA. Males and females aged 18-64 years (n 120). The quantities and measures were the most precise PSEA (lowest range of weights for estimated portion sizes). However, participants preferred household measures (e.g. 200 ml disposable cup) - deemed easy to use (median rating of 5), likely to use again in future (all scored either 4 or 5 on a scale from 1='not very likely' to 5='very likely to use again') and precise (narrow range of weights for estimated portion sizes). The majority indicated they would most likely use the PSEA preparing a meal (94 %), particularly dinner (86 %) in the home (89 %; all P<0·001) for amorphous grain foods. Household measures may be precise, easy to use and acceptable aids for estimating the appropriate portion size of amorphous grain foods.

Empirical likelihood inference in randomized clinical trials.

PubMed

Zhang, Biao

2017-01-01

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

Prevalence of Abuse Among Young Children with Rib Fractures: A Systematic Review

PubMed Central

Paine, Christine Weirich; Fakeye, Oludolapo; Christian, Cindy W.; Wood, Joanne N.

2016-01-01

Objectives We aimed to estimate the prevalence of abuse in young children presenting with rib fractures and to identify demographic, injury, and presentation-related characteristics that affect the probability that rib fractures are secondary to abuse. Methods We searched PubMed/MEDLINE and CINAHL databases for articles published in English between January 1, 1990 and June 30, 2014 on rib fracture etiology in children ≤ 5 years old. Two reviewers independently extracted predefined data elements and assigned quality ratings to included studies. Study-specific abuse prevalences and the sensitivities, specificities, and positive and negative likelihood ratios of patients’ demographic and clinical characteristics for abuse were calculated with 95% confidence intervals. Results Data for 1,396 children ≤ 48 months old with rib fractures were abstracted from 10 articles. Among infants < 12 months old, abuse prevalence ranged from 67% to 84%, whereas children 12-23 months old and 24-35 months old had study-specific abuse prevalences of 29% and 28% respectively. Age < 12 months was the only characteristic significantly associated with increased likelihood of abuse across multiple studies. Rib fracture location was not associated with likelihood of abuse. The retrospective design of the included studies and variations in ascertainment of cases, inclusion/exclusion criteria, and child abuse assessments prevented further meta-analysis. Conclusions Abuse is the most common cause of rib fractures in infants < 12 months old. Prospective studies with standardized methods are needed to improve accuracy in determining abuse prevalence among children with rib fractures and characteristics associated with abusive rib fractures. PMID:27749806

The contribution of passengers versus mobile phone use to motor vehicle crashes resulting in hospital attendance by the driver.

PubMed

McEvoy, Suzanne P; Stevenson, Mark R; Woodward, Mark

2007-11-01

There is evidence that mobile phone use while driving (including hands-free) is associated with motor vehicle crashes. However, whether the effects of mobile phone use differ from that of passengers in the vehicle remains unclear. The aim of this research was to estimate the risk of crash associated with passenger carriage and compare that with mobile phone use. A case-control study ('passenger study') was performed in Perth, Western Australia in 2003 and 2004. Cases were 274 drivers who attended hospital following a motor vehicle crash and controls were 1096 drivers (1:4 matching) recruited at service stations matched to the location and time and day of week of the crash. The results were compared with those of a case-crossover study ('mobile phone study') undertaken concurrently (n=456); 152 cases were common to both studies. Passenger carriage increased the likelihood of a crash (adjusted odds ratio (adj. OR), 95% confidence interval (95% CI), 1.6, 1.1-2.2). Drivers carrying two or more passengers were twice as likely to crash as unaccompanied drivers (adj. OR 2.2, 95% CI 1.3-3.8). By comparison, driver's use of a mobile phone within 5 min before a crash was associated with a fourfold increased likelihood of crashing (OR 4.1, 95% CI 2.2-7.7). Passenger carriage and increasing numbers of passengers are associated with an increased likelihood of crash, though not to the same extent as mobile phone use. Further research is needed to investigate the factors underlying the increased risks.

Development of advanced techniques for rotorcraft state estimation and parameter identification

NASA Technical Reports Server (NTRS)

Hall, W. E., Jr.; Bohn, J. G.; Vincent, J. H.

1980-01-01

An integrated methodology for rotorcraft system identification consists of rotorcraft mathematical modeling, three distinct data processing steps, and a technique for designing inputs to improve the identifiability of the data. These elements are as follows: (1) a Kalman filter smoother algorithm which estimates states and sensor errors from error corrupted data. Gust time histories and statistics may also be estimated; (2) a model structure estimation algorithm for isolating a model which adequately explains the data; (3) a maximum likelihood algorithm for estimating the parameters and estimates for the variance of these estimates; and (4) an input design algorithm, based on a maximum likelihood approach, which provides inputs to improve the accuracy of parameter estimates. Each step is discussed with examples to both flight and simulated data cases.

Inference of R 0 and Transmission Heterogeneity from the Size Distribution of Stuttering Chains

PubMed Central

Blumberg, Seth; Lloyd-Smith, James O.

2013-01-01

For many infectious disease processes such as emerging zoonoses and vaccine-preventable diseases, and infections occur as self-limited stuttering transmission chains. A mechanistic understanding of transmission is essential for characterizing the risk of emerging diseases and monitoring spatio-temporal dynamics. Thus methods for inferring and the degree of heterogeneity in transmission from stuttering chain data have important applications in disease surveillance and management. Previous researchers have used chain size distributions to infer , but estimation of the degree of individual-level variation in infectiousness (as quantified by the dispersion parameter, ) has typically required contact tracing data. Utilizing branching process theory along with a negative binomial offspring distribution, we demonstrate how maximum likelihood estimation can be applied to chain size data to infer both and the dispersion parameter that characterizes heterogeneity. While the maximum likelihood value for is a simple function of the average chain size, the associated confidence intervals are dependent on the inferred degree of transmission heterogeneity. As demonstrated for monkeypox data from the Democratic Republic of Congo, this impacts when a statistically significant change in is detectable. In addition, by allowing for superspreading events, inference of shifts the threshold above which a transmission chain should be considered anomalously large for a given value of (thus reducing the probability of false alarms about pathogen adaptation). Our analysis of monkeypox also clarifies the various ways that imperfect observation can impact inference of transmission parameters, and highlights the need to quantitatively evaluate whether observation is likely to significantly bias results. PMID:23658504

Environmental predictors and incubation period of AIDS-associated penicillium marneffei infection in Ho Chi Minh City, Vietnam.

PubMed

Bulterys, Philip L; Le, Thuy; Quang, Vo Minh; Nelson, Kenrad E; Lloyd-Smith, James O

2013-05-01

Penicillium marneffei is an emerging dimorphic mycosis endemic in Southeast Asia, and a leading cause of mortality among human immunodeficiency virus (HIV)-infected people in the region. Factors governing the seasonal incidence of P. marneffei infection are unknown, and may yield critical insights into possible reservoirs or modes of acquisition. This study included HIV-infected patients presenting with P. marneffei (n = 719) and Cryptococcus neoformans (n = 1598) infection to the Hospital for Tropical Diseases in Ho Chi Minh City, Vietnam, from 2004 to 2010, and temperature, humidity, wind, precipitation, and HIV-related admissions data for the corresponding period. We used multivariate regression modeling to identify factors associated with P. marneffei and C. neoformans admissions. We estimated the P. marneffei incubation period by considering profile likelihoods for different exposure-to-admission delays. We found that P. marneffei admissions were strongly associated with humidity (P < .001), and that precipitation, temperature, and wind did not add explanatory power. Cryptococcus neoformans admissions were not seasonal, and P. marneffei admissions were more common relative to C. neoformans admissions during months of high (≥85%) humidity (odds ratio, 1.49; 95% confidence interval [CI], 1.10-2.01). Maximum likelihood estimation suggested a P. marneffei incubation period of 1 week (95% CI, 0-3 weeks). Our findings suggest that humidity is the most important environmental predictor of P. marneffei admissions, and may drive exposure by facilitating fungal growth or spore release in the environment. In addition, it appears that a high proportion of penicilliosis patients present to the hospital with primary disseminated infection within 3 weeks of exposure.

Development and evaluation of modified envelope correlation method for deep tectonic tremor

NASA Astrophysics Data System (ADS)

Mizuno, N.; Ide, S.

2017-12-01

We develop a new location method for deep tectonic tremors, as an improvement of widely used envelope correlation method, and applied it to construct a tremor catalog in western Japan. Using the cross-correlation functions as objective functions and weighting components of data by the inverse of error variances, the envelope cross-correlation method is redefined as a maximum likelihood method. This method is also capable of multiple source detection, because when several events occur almost simultaneously, they appear as local maxima of likelihood.The average of weighted cross-correlation functions, defined as ACC, is a nonlinear function whose variable is a position of deep tectonic tremor. The optimization method has two steps. First, we fix the source depth to 30 km and use a grid search with 0.2 degree intervals to find the maxima of ACC, which are candidate event locations. Then, using each of the candidate locations as initial values, we apply a gradient method to determine horizontal and vertical components of a hypocenter. Sometimes, several source locations are determined in a time window of 5 minutes. We estimate the resolution, which is defined as a distance of sources to be detected separately by the location method, is about 100 km. The validity of this estimation is confirmed by a numerical test using synthetic waveforms. Applying to continuous seismograms in western Japan for over 10 years, the new method detected 27% more tremors than a previous method, owing to the multiple detection and improvement of accuracy by appropriate weighting scheme.

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).

Pseudo and conditional score approach to joint analysis of current count and current status data.

PubMed

Wen, Chi-Chung; Chen, Yi-Hau

2018-04-17

We develop a joint analysis approach for recurrent and nonrecurrent event processes subject to case I interval censorship, which are also known in literature as current count and current status data, respectively. We use a shared frailty to link the recurrent and nonrecurrent event processes, while leaving the distribution of the frailty fully unspecified. Conditional on the frailty, the recurrent event is assumed to follow a nonhomogeneous Poisson process, and the mean function of the recurrent event and the survival function of the nonrecurrent event are assumed to follow some general form of semiparametric transformation models. Estimation of the models is based on the pseudo-likelihood and the conditional score techniques. The resulting estimators for the regression parameters and the unspecified baseline functions are shown to be consistent with rates of square and cubic roots of the sample size, respectively. Asymptotic normality with closed-form asymptotic variance is derived for the estimator of the regression parameters. We apply the proposed method to a fracture-osteoporosis survey data to identify risk factors jointly for fracture and osteoporosis in elders, while accounting for association between the two events within a subject. © 2018, The International Biometric Society.

On the Computation of the RMSEA and CFI from the Mean-And-Variance Corrected Test Statistic with Nonnormal Data in SEM.

PubMed

Savalei, Victoria

2018-01-01

A new type of nonnormality correction to the RMSEA has recently been developed, which has several advantages over existing corrections. In particular, the new correction adjusts the sample estimate of the RMSEA for the inflation due to nonnormality, while leaving its population value unchanged, so that established cutoff criteria can still be used to judge the degree of approximate fit. A confidence interval (CI) for the new robust RMSEA based on the mean-corrected ("Satorra-Bentler") test statistic has also been proposed. Follow up work has provided the same type of nonnormality correction for the CFI (Brosseau-Liard & Savalei, 2014). These developments have recently been implemented in lavaan. This note has three goals: a) to show how to compute the new robust RMSEA and CFI from the mean-and-variance corrected test statistic; b) to offer a new CI for the robust RMSEA based on the mean-and-variance corrected test statistic; and c) to caution that the logic of the new nonnormality corrections to RMSEA and CFI is most appropriate for the maximum likelihood (ML) estimator, and cannot easily be generalized to the most commonly used categorical data estimators.

Robust analysis of semiparametric renewal process models

PubMed Central

Lin, Feng-Chang; Truong, Young K.; Fine, Jason P.

2013-01-01

Summary A rate model is proposed for a modulated renewal process comprising a single long sequence, where the covariate process may not capture the dependencies in the sequence as in standard intensity models. We consider partial likelihood-based inferences under a semiparametric multiplicative rate model, which has been widely studied in the context of independent and identical data. Under an intensity model, gap times in a single long sequence may be used naively in the partial likelihood with variance estimation utilizing the observed information matrix. Under a rate model, the gap times cannot be treated as independent and studying the partial likelihood is much more challenging. We employ a mixing condition in the application of limit theory for stationary sequences to obtain consistency and asymptotic normality. The estimator's variance is quite complicated owing to the unknown gap times dependence structure. We adapt block bootstrapping and cluster variance estimators to the partial likelihood. Simulation studies and an analysis of a semiparametric extension of a popular model for neural spike train data demonstrate the practical utility of the rate approach in comparison with the intensity approach. PMID:24550568

Optimal time points sampling in pathway modelling.

PubMed

Hu, Shiyan

2004-01-01

Modelling cellular dynamics based on experimental data is at the heart of system biology. Considerable progress has been made to dynamic pathway modelling as well as the related parameter estimation. However, few of them gives consideration for the issue of optimal sampling time selection for parameter estimation. Time course experiments in molecular biology rarely produce large and accurate data sets and the experiments involved are usually time consuming and expensive. Therefore, to approximate parameters for models with only few available sampling data is of significant practical value. For signal transduction, the sampling intervals are usually not evenly distributed and are based on heuristics. In the paper, we investigate an approach to guide the process of selecting time points in an optimal way to minimize the variance of parameter estimates. In the method, we first formulate the problem to a nonlinear constrained optimization problem by maximum likelihood estimation. We then modify and apply a quantum-inspired evolutionary algorithm, which combines the advantages of both quantum computing and evolutionary computing, to solve the optimization problem. The new algorithm does not suffer from the morass of selecting good initial values and being stuck into local optimum as usually accompanied with the conventional numerical optimization techniques. The simulation results indicate the soundness of the new method.

Effect of inclusion or non-inclusion of short lactations and cow and/or dam genetic group on genetic evaluation of Girolando dairy cattle.

PubMed

Canaza-Cayo, A W; Silva, M V G B; Cobuci, J A; Martins, M F; Lopes, P S

2016-04-04

The objective of this study was to evaluate the effects of inclusion or non-inclusion of short lactations and cow (CGG) and/or dam (DGG) genetic group on the genetic evaluation of 305-day milk yield (MY305), age at first calving (AFC), and first calving interval (FCI) of Girolando cows. Covariance components were estimated by the restricted maximum likelihood method in an animal model of single trait analyses. The heritability estimates for MY305, AFC, and FCI ranged from 0.23 to 0.29, 0.40 to 0.44, and 0.13 to 0.14, respectively, when short lactations were not included, and from 0.23 to 0.28, 0.39 to 0.43, and 0.13 to 0.14, respectively, when short lactations were included. The inclusion of short lactations caused little variation in the variance components and heritability estimates of traits, but their non-inclusion resulted in the re-ranking of animals. Models with CGG or DGG fixed effects had higher heritability estimates for all traits compared with models that consider these two effects simultaneously. We recommend using the model with fixed effects of CGG and inclusion of short lactations for the genetic evaluation of Girolando cattle.

Crash test rating and likelihood of major thoracoabdominal injury in motor vehicle crashes: the new car assessment program side-impact crash test, 1998-2010.

PubMed

Figler, Bradley D; Mack, Christopher D; Kaufman, Robert; Wessells, Hunter; Bulger, Eileen; Smith, Thomas G; Voelzke, Bryan

2014-03-01

The National Highway Traffic Safety Administration's New Car Assessment Program (NCAP) implemented side-impact crash testing on all new vehicles since 1998 to assess the likelihood of major thoracoabdominal injuries during a side-impact crash. Higher crash test rating is intended to indicate a safer car, but the real-world applicability of these ratings is unknown. Our objective was to determine the relationship between a vehicle's NCAP side-impact crash test rating and the risk of major thoracoabdominal injury among the vehicle's occupants in real-world side-impact motor vehicle crashes. The National Automotive Sampling System Crashworthiness Data System contains detailed crash and injury data in a sample of major crashes in the United States. For model years 1998 to 2010 and crash years 1999 to 2010, 68,124 occupants were identified in the Crashworthiness Data System database. Because 47% of cases were missing crash severity (ΔV), multiple imputation was used to estimate the missing values. The primary predictor of interest was the occupant vehicle's NCAP side-impact crash test rating, and the outcome of interest was the presence of major (Abbreviated Injury Scale [AIS] score ≥ 3) thoracoabdominal injury. In multivariate analysis, increasing NCAP crash test rating was associated with lower likelihood of major thoracoabdominal injury at high (odds ratio [OR], 0.8; 95% confidence interval [CI], 0.7-0.9; p < 0.01) and medium (OR, 0.9; 95% CI, 0.8-1.0; p < 0.05) crash severity (ΔV), but not at low ΔV (OR, 0.95; 95% CI, 0.8-1.2; p = 0.55). In our model, older age and absence of seat belt use were associated with greater likelihood of major thoracoabdominal injury at low and medium ΔV (p < 0.001), but not at high ΔV (p ≥ 0.09). Among adults in model year 1998 to 2010 vehicles involved in medium and high severity motor vehicle crashes, a higher NCAP side-impact crash test rating is associated with a lower likelihood of major thoracoabdominal trauma. Epidemiologic study, level III.

Efficient estimators for likelihood ratio sensitivity indices of complex stochastic dynamics.

PubMed

Arampatzis, Georgios; Katsoulakis, Markos A; Rey-Bellet, Luc

2016-03-14

We demonstrate that centered likelihood ratio estimators for the sensitivity indices of complex stochastic dynamics are highly efficient with low, constant in time variance and consequently they are suitable for sensitivity analysis in long-time and steady-state regimes. These estimators rely on a new covariance formulation of the likelihood ratio that includes as a submatrix a Fisher information matrix for stochastic dynamics and can also be used for fast screening of insensitive parameters and parameter combinations. The proposed methods are applicable to broad classes of stochastic dynamics such as chemical reaction networks, Langevin-type equations and stochastic models in finance, including systems with a high dimensional parameter space and/or disparate decorrelation times between different observables. Furthermore, they are simple to implement as a standard observable in any existing simulation algorithm without additional modifications.

A single-index threshold Cox proportional hazard model for identifying a treatment-sensitive subset based on multiple biomarkers.

PubMed

He, Ye; Lin, Huazhen; Tu, Dongsheng

2018-06-04

In this paper, we introduce a single-index threshold Cox proportional hazard model to select and combine biomarkers to identify patients who may be sensitive to a specific treatment. A penalized smoothed partial likelihood is proposed to estimate the parameters in the model. A simple, efficient, and unified algorithm is presented to maximize this likelihood function. The estimators based on this likelihood function are shown to be consistent and asymptotically normal. Under mild conditions, the proposed estimators also achieve the oracle property. The proposed approach is evaluated through simulation analyses and application to the analysis of data from two clinical trials, one involving patients with locally advanced or metastatic pancreatic cancer and one involving patients with resectable lung cancer. Copyright © 2018 John Wiley & Sons, Ltd.

Efficient estimators for likelihood ratio sensitivity indices of complex stochastic dynamics

NASA Astrophysics Data System (ADS)

Arampatzis, Georgios; Katsoulakis, Markos A.; Rey-Bellet, Luc

2016-03-01

We demonstrate that centered likelihood ratio estimators for the sensitivity indices of complex stochastic dynamics are highly efficient with low, constant in time variance and consequently they are suitable for sensitivity analysis in long-time and steady-state regimes. These estimators rely on a new covariance formulation of the likelihood ratio that includes as a submatrix a Fisher information matrix for stochastic dynamics and can also be used for fast screening of insensitive parameters and parameter combinations. The proposed methods are applicable to broad classes of stochastic dynamics such as chemical reaction networks, Langevin-type equations and stochastic models in finance, including systems with a high dimensional parameter space and/or disparate decorrelation times between different observables. Furthermore, they are simple to implement as a standard observable in any existing simulation algorithm without additional modifications.

Efficient estimators for likelihood ratio sensitivity indices of complex stochastic dynamics

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

Arampatzis, Georgios; Katsoulakis, Markos A.; Rey-Bellet, Luc

2016-03-14

We demonstrate that centered likelihood ratio estimators for the sensitivity indices of complex stochastic dynamics are highly efficient with low, constant in time variance and consequently they are suitable for sensitivity analysis in long-time and steady-state regimes. These estimators rely on a new covariance formulation of the likelihood ratio that includes as a submatrix a Fisher information matrix for stochastic dynamics and can also be used for fast screening of insensitive parameters and parameter combinations. The proposed methods are applicable to broad classes of stochastic dynamics such as chemical reaction networks, Langevin-type equations and stochastic models in finance, including systemsmore » with a high dimensional parameter space and/or disparate decorrelation times between different observables. Furthermore, they are simple to implement as a standard observable in any existing simulation algorithm without additional modifications.« less

Approximate likelihood approaches for detecting the influence of primordial gravitational waves in cosmic microwave background polarization

NASA Astrophysics Data System (ADS)

Pan, Zhen; Anderes, Ethan; Knox, Lloyd

2018-05-01

One of the major targets for next-generation cosmic microwave background (CMB) experiments is the detection of the primordial B-mode signal. Planning is under way for Stage-IV experiments that are projected to have instrumental noise small enough to make lensing and foregrounds the dominant source of uncertainty for estimating the tensor-to-scalar ratio r from polarization maps. This makes delensing a crucial part of future CMB polarization science. In this paper we present a likelihood method for estimating the tensor-to-scalar ratio r from CMB polarization observations, which combines the benefits of a full-scale likelihood approach with the tractability of the quadratic delensing technique. This method is a pixel space, all order likelihood analysis of the quadratic delensed B modes, and it essentially builds upon the quadratic delenser by taking into account all order lensing and pixel space anomalies. Its tractability relies on a crucial factorization of the pixel space covariance matrix of the polarization observations which allows one to compute the full Gaussian approximate likelihood profile, as a function of r , at the same computational cost of a single likelihood evaluation.

Effect of formal and informal likelihood functions on uncertainty assessment in a single event rainfall-runoff model

NASA Astrophysics Data System (ADS)

Nourali, Mahrouz; Ghahraman, Bijan; Pourreza-Bilondi, Mohsen; Davary, Kamran

2016-09-01

In the present study, DREAM(ZS), Differential Evolution Adaptive Metropolis combined with both formal and informal likelihood functions, is used to investigate uncertainty of parameters of the HEC-HMS model in Tamar watershed, Golestan province, Iran. In order to assess the uncertainty of 24 parameters used in HMS, three flood events were used to calibrate and one flood event was used to validate the posterior distributions. Moreover, performance of seven different likelihood functions (L1-L7) was assessed by means of DREAM(ZS)approach. Four likelihood functions, L1-L4, Nash-Sutcliffe (NS) efficiency, Normalized absolute error (NAE), Index of agreement (IOA), and Chiew-McMahon efficiency (CM), is considered as informal, whereas remaining (L5-L7) is represented in formal category. L5 focuses on the relationship between the traditional least squares fitting and the Bayesian inference, and L6, is a hetereoscedastic maximum likelihood error (HMLE) estimator. Finally, in likelihood function L7, serial dependence of residual errors is accounted using a first-order autoregressive (AR) model of the residuals. According to the results, sensitivities of the parameters strongly depend on the likelihood function, and vary for different likelihood functions. Most of the parameters were better defined by formal likelihood functions L5 and L7 and showed a high sensitivity to model performance. Posterior cumulative distributions corresponding to the informal likelihood functions L1, L2, L3, L4 and the formal likelihood function L6 are approximately the same for most of the sub-basins, and these likelihood functions depict almost a similar effect on sensitivity of parameters. 95% total prediction uncertainty bounds bracketed most of the observed data. Considering all the statistical indicators and criteria of uncertainty assessment, including RMSE, KGE, NS, P-factor and R-factor, results showed that DREAM(ZS) algorithm performed better under formal likelihood functions L5 and L7, but likelihood function L5 may result in biased and unreliable estimation of parameters due to violation of the residualerror assumptions. Thus, likelihood function L7 provides posterior distribution of model parameters credibly and therefore can be employed for further applications.

The effect of high leverage points on the logistic ridge regression estimator having multicollinearity

NASA Astrophysics Data System (ADS)

Ariffin, Syaiba Balqish; Midi, Habshah

2014-06-01

This article is concerned with the performance of logistic ridge regression estimation technique in the presence of multicollinearity and high leverage points. In logistic regression, multicollinearity exists among predictors and in the information matrix. The maximum likelihood estimator suffers a huge setback in the presence of multicollinearity which cause regression estimates to have unduly large standard errors. To remedy this problem, a logistic ridge regression estimator is put forward. It is evident that the logistic ridge regression estimator outperforms the maximum likelihood approach for handling multicollinearity. The effect of high leverage points are then investigated on the performance of the logistic ridge regression estimator through real data set and simulation study. The findings signify that logistic ridge regression estimator fails to provide better parameter estimates in the presence of both high leverage points and multicollinearity.

Estimation of the Arrival Time and Duration of a Radio Signal with Unknown Amplitude and Initial Phase

NASA Astrophysics Data System (ADS)

Trifonov, A. P.; Korchagin, Yu. E.; Korol'kov, S. V.

2018-05-01

We synthesize the quasi-likelihood, maximum-likelihood, and quasioptimal algorithms for estimating the arrival time and duration of a radio signal with unknown amplitude and initial phase. The discrepancies between the hardware and software realizations of the estimation algorithm are shown. The characteristics of the synthesized-algorithm operation efficiency are obtained. Asymptotic expressions for the biases, variances, and the correlation coefficient of the arrival-time and duration estimates, which hold true for large signal-to-noise ratios, are derived. The accuracy losses of the estimates of the radio-signal arrival time and duration because of the a priori ignorance of the amplitude and initial phase are determined.

Relationship of Antibiotic Treatment to Recovery after Acute FEV1 Decline in Children with Cystic Fibrosis.

PubMed

Morgan, Wayne J; Wagener, Jeffrey S; Pasta, David J; Millar, Stefanie J; VanDevanter, Donald R; Konstan, Michael W

2017-06-01

Children with cystic fibrosis often experience acute declines in lung function. We previously showed that such declines are not always treated with antibiotics, but we did not assess whether treatment improves the likelihood of recovery. To determine whether new antibiotic treatment was associated with recovery from acute FEV 1 decline. We studied episodes of FEV 1 decline (≥10% from baseline) in the Epidemiologic Study of Cystic Fibrosis. Treatments were hospitalization, home intravenous antibiotic, new inhaled oral quinolone, or other oral antibiotic. We used logistic regression to evaluate whether treatment was associated with recovery to baseline or near baseline. Logistic regression of 9,875 patients showed that new antibiotic treatment was associated with an increased likelihood of recovery to 90% of baseline (P < 0.001), especially for hospitalization compared with no new antibiotic (odds ratio [OR], 2.79; 95% confidence interval, 2.41-3.23). All four outpatient treatments were associated with greater likelihood of recovery compared with no treatment (OR, 1.27-1.64). Inpatient treatment was better than outpatient treatment (OR, 1.94; 95% confidence interval, 1.68-2.23). Treatment-type ORs were similar across recovery criteria and levels of baseline lung function. New antibiotic therapy, and especially inpatient treatment, is associated with greater likelihood of recovery after acute decline in FEV 1 . Benefits extend across all disease stages and are especially important in patients with high lung function, who are at greatest risk for FEV 1 decline.

Species delimitation using Bayes factors: simulations and application to the Sceloporus scalaris species group (Squamata: Phrynosomatidae).

PubMed

Grummer, Jared A; Bryson, Robert W; Reeder, Tod W

2014-03-01

Current molecular methods of species delimitation are limited by the types of species delimitation models and scenarios that can be tested. Bayes factors allow for more flexibility in testing non-nested species delimitation models and hypotheses of individual assignment to alternative lineages. Here, we examined the efficacy of Bayes factors in delimiting species through simulations and empirical data from the Sceloporus scalaris species group. Marginal-likelihood scores of competing species delimitation models, from which Bayes factor values were compared, were estimated with four different methods: harmonic mean estimation (HME), smoothed harmonic mean estimation (sHME), path-sampling/thermodynamic integration (PS), and stepping-stone (SS) analysis. We also performed model selection using a posterior simulation-based analog of the Akaike information criterion through Markov chain Monte Carlo analysis (AICM). Bayes factor species delimitation results from the empirical data were then compared with results from the reversible-jump MCMC (rjMCMC) coalescent-based species delimitation method Bayesian Phylogenetics and Phylogeography (BP&P). Simulation results show that HME and sHME perform poorly compared with PS and SS marginal-likelihood estimators when identifying the true species delimitation model. Furthermore, Bayes factor delimitation (BFD) of species showed improved performance when species limits are tested by reassigning individuals between species, as opposed to either lumping or splitting lineages. In the empirical data, BFD through PS and SS analyses, as well as the rjMCMC method, each provide support for the recognition of all scalaris group taxa as independent evolutionary lineages. Bayes factor species delimitation and BP&P also support the recognition of three previously undescribed lineages. In both simulated and empirical data sets, harmonic and smoothed harmonic mean marginal-likelihood estimators provided much higher marginal-likelihood estimates than PS and SS estimators. The AICM displayed poor repeatability in both simulated and empirical data sets, and produced inconsistent model rankings across replicate runs with the empirical data. Our results suggest that species delimitation through the use of Bayes factors with marginal-likelihood estimates via PS or SS analyses provide a useful and complementary alternative to existing species delimitation methods.

Estimating population genetic parameters and comparing model goodness-of-fit using DNA sequences with error

PubMed Central

Liu, Xiaoming; Fu, Yun-Xin; Maxwell, Taylor J.; Boerwinkle, Eric

2010-01-01

It is known that sequencing error can bias estimation of evolutionary or population genetic parameters. This problem is more prominent in deep resequencing studies because of their large sample size n, and a higher probability of error at each nucleotide site. We propose a new method based on the composite likelihood of the observed SNP configurations to infer population mutation rate θ = 4Neμ, population exponential growth rate R, and error rate ɛ, simultaneously. Using simulation, we show the combined effects of the parameters, θ, n, ɛ, and R on the accuracy of parameter estimation. We compared our maximum composite likelihood estimator (MCLE) of θ with other θ estimators that take into account the error. The results show the MCLE performs well when the sample size is large or the error rate is high. Using parametric bootstrap, composite likelihood can also be used as a statistic for testing the model goodness-of-fit of the observed DNA sequences. The MCLE method is applied to sequence data on the ANGPTL4 gene in 1832 African American and 1045 European American individuals. PMID:19952140

INFERRING THE ECCENTRICITY DISTRIBUTION

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

Hogg, David W.; Bovy, Jo; Myers, Adam D., E-mail: david.hogg@nyu.ed

2010-12-20

Standard maximum-likelihood estimators for binary-star and exoplanet eccentricities are biased high, in the sense that the estimated eccentricity tends to be larger than the true eccentricity. As with most non-trivial observables, a simple histogram of estimated eccentricities is not a good estimate of the true eccentricity distribution. Here, we develop and test a hierarchical probabilistic method for performing the relevant meta-analysis, that is, inferring the true eccentricity distribution, taking as input the likelihood functions for the individual star eccentricities, or samplings of the posterior probability distributions for the eccentricities (under a given, uninformative prior). The method is a simple implementationmore » of a hierarchical Bayesian model; it can also be seen as a kind of heteroscedastic deconvolution. It can be applied to any quantity measured with finite precision-other orbital parameters, or indeed any astronomical measurements of any kind, including magnitudes, distances, or photometric redshifts-so long as the measurements have been communicated as a likelihood function or a posterior sampling.« less

Precision Parameter Estimation and Machine Learning

NASA Astrophysics Data System (ADS)

Wandelt, Benjamin D.

2008-12-01

I discuss the strategy of ``Acceleration by Parallel Precomputation and Learning'' (AP-PLe) that can vastly accelerate parameter estimation in high-dimensional parameter spaces and costly likelihood functions, using trivially parallel computing to speed up sequential exploration of parameter space. This strategy combines the power of distributed computing with machine learning and Markov-Chain Monte Carlo techniques efficiently to explore a likelihood function, posterior distribution or χ2-surface. This strategy is particularly successful in cases where computing the likelihood is costly and the number of parameters is moderate or large. We apply this technique to two central problems in cosmology: the solution of the cosmological parameter estimation problem with sufficient accuracy for the Planck data using PICo; and the detailed calculation of cosmological helium and hydrogen recombination with RICO. Since the APPLe approach is designed to be able to use massively parallel resources to speed up problems that are inherently serial, we can bring the power of distributed computing to bear on parameter estimation problems. We have demonstrated this with the CosmologyatHome project.

Methods for estimating drought streamflow probabilities for Virginia streams

USGS Publications Warehouse

Austin, Samuel H.

2014-01-01

Maximum likelihood logistic regression model equations used to estimate drought flow probabilities for Virginia streams are presented for 259 hydrologic basins in Virginia. Winter streamflows were used to estimate the likelihood of streamflows during the subsequent drought-prone summer months. The maximum likelihood logistic regression models identify probable streamflows from 5 to 8 months in advance. More than 5 million streamflow daily values collected over the period of record (January 1, 1900 through May 16, 2012) were compiled and analyzed over a minimum 10-year (maximum 112-year) period of record. The analysis yielded the 46,704 equations with statistically significant fit statistics and parameter ranges published in two tables in this report. These model equations produce summer month (July, August, and September) drought flow threshold probabilities as a function of streamflows during the previous winter months (November, December, January, and February). Example calculations are provided, demonstrating how to use the equations to estimate probable streamflows as much as 8 months in advance.

Is the first seizure epilepsy--and when?

PubMed

Lawn, Nicholas; Chan, Josephine; Lee, Judy; Dunne, John

2015-09-01

Epilepsy has recently been redefined to include a single unprovoked seizure if the probability of recurrence is ≥60% over the following 10 years. This definition is based on the estimated risk of a third seizure after two unprovoked seizures, using the lower-limit 95% confidence interval (CI) at 4 years, and does not account for the initially high recurrence rate after first-ever seizure that rapidly falls with increasing duration of seizure freedom. We analyzed long-term outcomes after the first-ever seizure, and the influence of duration of seizure freedom on the likelihood of seizure recurrence, and their relevance to the new definition of epilepsy. Prospective analysis of 798 adults with a first-ever unprovoked seizure seen at a hospital-based first seizure clinic between 2000 and 2011. The likelihood of seizure recurrence was analyzed according to the duration of seizure freedom, etiology, electroencephalography (EEG), and neuroimaging findings. The likelihood of seizure recurrence at 10 years was ≥60% in patients with epileptiform abnormalities on EEG or neuroimaging abnormalities, therefore, meeting the new definition of epilepsy. However, the risk of recurrence was highly time dependent; after a brief period (≤12 weeks) of seizure freedom, no patient group continued to fulfill the new definition of epilepsy. Of 407 patients who had a second seizure, the likelihood of a third seizure at 4 years was 68% (95% CI 63-73%) and at 10 years was 85% (95% CI 79-91%). The duration of seizure freedom following first-ever seizure substantially influences the risk of recurrence, with none of our patients fulfilling the new definition of epilepsy after a short period of seizure freedom. When a threshold was applied based on the 10-year risk of a third seizure from our data, no first-seizure patient group ever had epilepsy. These data may be utilized in a definition of epilepsy after a first-ever seizure. Wiley Periodicals, Inc. © 2015 International League Against Epilepsy.

Trends in the Prescription and Long-Term Utilization of Antidementia Drugs Among Patients with Alzheimer's Disease in Spain: A Cohort Study Using the Registry of Dementias of Girona.

PubMed

Calvó-Perxas, Laia; Turró-Garriga, Oriol; Vilalta-Franch, Joan; Lozano-Gallego, Manuela; de Eugenio, Rosa; Márquez, Fabián; Carmona, Olga; Gich, Jordi; Manzano, Anna; Viñas, Marta; Roig, Anna Mª; Garre-Olmo, Josep

2017-04-01

Acetylcholinesterase inhibitors (AChEIs) and the N-methyl D-aspartate-antagonist memantine are indicated for the symptomatic treatment of Alzheimer's disease (AD). Our aims were to describe the baseline characteristics of patients with AD according to prescription of these treatments after the diagnostic work-up to describe long-term trends in the use of these medications and to identify baseline characteristics associated with the frequency of use of each treatment. This was a cohort study with a sample of 2992 patients with AD recorded in the Registry of Dementias of Girona (ReDeGi) between 2007 and 2014. Consumption of AChEIs and memantine was assessed using the Pharmacy Unit database from the Public Catalan Healthcare Service. We used generalized estimating equation analyses to identify the baseline characteristics associated with the consumption of AChEIs and memantine over time. Most of the patients (70.4%; 95% confidence interval [CI] 68.7-72.0) were prescribed antidementia medication at the time of diagnosis. Of these, 75.0% (95% CI 73.1-76.8) were prescribed AChEIs, 14.7% (95% CI 13.2-16.3) were prescribed an AChEI plus memantine, and 10.3% (95% CI 9.0-11.6) were prescribed memantine. Advanced age reduced the likelihood of AChEI consumption. Mild dementia severity increased the use of AChEIs, and moderate-advanced dementia increased the likelihood of memantine consumption. After diagnosis, the likelihood of AChEI consumption decreased from the first year until the fifth, whereas the likelihood of memantine consumption, either alone or in combination with AChEIs, increased. Antidementia drug use in this study showed the initial use of AChEIs alone with later use of AChEIs in combination with memantine and memantine alone in older patients with severe AD. Our findings are in agreement with current clinical practice guidelines for the pharmacological treatment of AD.

Kumaraswamy autoregressive moving average models for double bounded environmental data

NASA Astrophysics Data System (ADS)

Bayer, Fábio Mariano; Bayer, Débora Missio; Pumi, Guilherme

2017-12-01

In this paper we introduce the Kumaraswamy autoregressive moving average models (KARMA), which is a dynamic class of models for time series taking values in the double bounded interval (a,b) following the Kumaraswamy distribution. The Kumaraswamy family of distribution is widely applied in many areas, especially hydrology and related fields. Classical examples are time series representing rates and proportions observed over time. In the proposed KARMA model, the median is modeled by a dynamic structure containing autoregressive and moving average terms, time-varying regressors, unknown parameters and a link function. We introduce the new class of models and discuss conditional maximum likelihood estimation, hypothesis testing inference, diagnostic analysis and forecasting. In particular, we provide closed-form expressions for the conditional score vector and conditional Fisher information matrix. An application to environmental real data is presented and discussed.

Massive optimal data compression and density estimation for scalable, likelihood-free inference in cosmology

NASA Astrophysics Data System (ADS)

Alsing, Justin; Wandelt, Benjamin; Feeney, Stephen

2018-07-01

Many statistical models in cosmology can be simulated forwards but have intractable likelihood functions. Likelihood-free inference methods allow us to perform Bayesian inference from these models using only forward simulations, free from any likelihood assumptions or approximations. Likelihood-free inference generically involves simulating mock data and comparing to the observed data; this comparison in data space suffers from the curse of dimensionality and requires compression of the data to a small number of summary statistics to be tractable. In this paper, we use massive asymptotically optimal data compression to reduce the dimensionality of the data space to just one number per parameter, providing a natural and optimal framework for summary statistic choice for likelihood-free inference. Secondly, we present the first cosmological application of Density Estimation Likelihood-Free Inference (DELFI), which learns a parametrized model for joint distribution of data and parameters, yielding both the parameter posterior and the model evidence. This approach is conceptually simple, requires less tuning than traditional Approximate Bayesian Computation approaches to likelihood-free inference and can give high-fidelity posteriors from orders of magnitude fewer forward simulations. As an additional bonus, it enables parameter inference and Bayesian model comparison simultaneously. We demonstrate DELFI with massive data compression on an analysis of the joint light-curve analysis supernova data, as a simple validation case study. We show that high-fidelity posterior inference is possible for full-scale cosmological data analyses with as few as ˜104 simulations, with substantial scope for further improvement, demonstrating the scalability of likelihood-free inference to large and complex cosmological data sets.

Effects of time-shifted data on flight determined stability and control derivatives

NASA Technical Reports Server (NTRS)

Steers, S. T.; Iliff, K. W.

1975-01-01

Flight data were shifted in time by various increments to assess the effects of time shifts on estimates of stability and control derivatives produced by a maximum likelihood estimation method. Derivatives could be extracted from flight data with the maximum likelihood estimation method even if there was a considerable time shift in the data. Time shifts degraded the estimates of the derivatives, but the degradation was in a consistent rather than a random pattern. Time shifts in the control variables caused the most degradation, and the lateral-directional rotary derivatives were affected the most by time shifts in any variable.

Beyond valence in the perception of likelihood: the role of emotion specificity.

PubMed

DeSteno, D; Petty, R E; Wegener, D T; Rucker, D D

2000-03-01

Positive and negative moods have been shown to increase likelihood estimates of future events matching these states in valence (e.g., E. J. Johnson & A. Tversky, 1983). In the present article, 4 studies provide evidence that this congruency bias (a) is not limited to valence but functions in an emotion-specific manner, (b) derives from the informational value of emotions, and (c) is not the inevitable outcome of likelihood assessment under heightened emotion. Specifically, Study 1 demonstrates that sadness and anger, 2 distinct, negative emotions, differentially bias likelihood estimates of sad and angering events. Studies 2 and 3 replicate this finding in addition to supporting an emotion-as-information (cf. N. Schwarz & G. L. Clore, 1983), as opposed to a memory-based, mediating process for the bias. Finally, Study 4 shows that when the source of the emotion is salient, a reversal of the bias can occur given greater cognitive effort aimed at accuracy.

Proactive Control: Neural Oscillatory Correlates of Conflict Anticipation and Response Slowing.

PubMed

Chang, Andrew; Ide, Jaime S; Li, Hsin-Hung; Chen, Chien-Chung; Li, Chiang-Shan R

2017-01-01

Proactive control allows us to anticipate environmental changes and adjust behavioral strategy. In the laboratory, investigators have used a number of different behavioral paradigms, including the stop-signal task (SST), to examine the neural processes of proactive control. Previous functional MRI studies of the SST have demonstrated regional responses to conflict anticipation-the likelihood of a stop signal or P(stop) as estimated by a Bayesian model-and reaction time (RT) slowing and how these responses are interrelated. Here, in an electrophysiological study, we investigated the time-frequency domain substrates of proactive control. The results showed that conflict anticipation as indexed by P(stop) was positively correlated with the power in low-theta band (3-5 Hz) in the fixation (trial onset)-locked interval, and go-RT was negatively correlated with the power in delta-theta band (2-8 Hz) in the go-locked interval. Stimulus prediction error was positively correlated with the power in the low-beta band (12-22 Hz) in the stop-locked interval. Further, the power of the P(stop) and go-RT clusters was negatively correlated, providing a mechanism relating conflict anticipation to RT slowing in the SST. Source reconstruction with beamformer localized these time-frequency activities close to brain regions as revealed by functional MRI in earlier work. These are the novel results to show oscillatory electrophysiological substrates in support of trial-by-trial behavioral adjustment for proactive control.

Pointwise nonparametric maximum likelihood estimator of stochastically ordered survivor functions

PubMed Central

Park, Yongseok; Taylor, Jeremy M. G.; Kalbfleisch, John D.

2012-01-01

In this paper, we consider estimation of survivor functions from groups of observations with right-censored data when the groups are subject to a stochastic ordering constraint. Many methods and algorithms have been proposed to estimate distribution functions under such restrictions, but none have completely satisfactory properties when the observations are censored. We propose a pointwise constrained nonparametric maximum likelihood estimator, which is defined at each time t by the estimates of the survivor functions subject to constraints applied at time t only. We also propose an efficient method to obtain the estimator. The estimator of each constrained survivor function is shown to be nonincreasing in t, and its consistency and asymptotic distribution are established. A simulation study suggests better small and large sample properties than for alternative estimators. An example using prostate cancer data illustrates the method. PMID:23843661

Closed-loop carrier phase synchronization techniques motivated by likelihood functions

NASA Technical Reports Server (NTRS)

Tsou, H.; Hinedi, S.; Simon, M.

1994-01-01

This article reexamines the notion of closed-loop carrier phase synchronization motivated by the theory of maximum a posteriori phase estimation with emphasis on the development of new structures based on both maximum-likelihood and average-likelihood functions. The criterion of performance used for comparison of all the closed-loop structures discussed is the mean-squared phase error for a fixed-loop bandwidth.

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.

Approximate likelihood calculation on a phylogeny for Bayesian estimation of divergence times.

PubMed

dos Reis, Mario; Yang, Ziheng

2011-07-01

The molecular clock provides a powerful way to estimate species divergence times. If information on some species divergence times is available from the fossil or geological record, it can be used to calibrate a phylogeny and estimate divergence times for all nodes in the tree. The Bayesian method provides a natural framework to incorporate different sources of information concerning divergence times, such as information in the fossil and molecular data. Current models of sequence evolution are intractable in a Bayesian setting, and Markov chain Monte Carlo (MCMC) is used to generate the posterior distribution of divergence times and evolutionary rates. This method is computationally expensive, as it involves the repeated calculation of the likelihood function. Here, we explore the use of Taylor expansion to approximate the likelihood during MCMC iteration. The approximation is much faster than conventional likelihood calculation. However, the approximation is expected to be poor when the proposed parameters are far from the likelihood peak. We explore the use of parameter transforms (square root, logarithm, and arcsine) to improve the approximation to the likelihood curve. We found that the new methods, particularly the arcsine-based transform, provided very good approximations under relaxed clock models and also under the global clock model when the global clock is not seriously violated. The approximation is poorer for analysis under the global clock when the global clock is seriously wrong and should thus not be used. The results suggest that the approximate method may be useful for Bayesian dating analysis using large data sets.

Computing and software

USGS Publications Warehouse

White, Gary C.; Hines, J.E.

2004-01-01

The reality is that the statistical methods used for analysis of data depend upon the availability of software. Analysis of marked animal data is no different than the rest of the statistical field. The methods used for analysis are those that are available in reliable software packages. Thus, the critical importance of having reliable, up–to–date software available to biologists is obvious. Statisticians have continued to develop more robust models, ever expanding the suite of potential analysis methodsavailable. But without software to implement these newer methods, they will languish in the abstract, and not be applied to the problems deserving them.In the Computers and Software Session, two new software packages are described, a comparison of implementation of methods for the estimation of nest survival is provided, and a more speculative paper about how the next generation of software might be structured is presented.Rotella et al. (2004) compare nest survival estimation with different software packages: SAS logistic regression, SAS non–linear mixed models, and Program MARK. Nests are assumed to be visited at various, possibly infrequent, intervals. All of the approaches described compute nest survival with the same likelihood, and require that the age of the nest is known to account for nests that eventually hatch. However, each approach offers advantages and disadvantages, explored by Rotella et al. (2004).Efford et al. (2004) present a new software package called DENSITY. The package computes population abundance and density from trapping arrays and other detection methods with a new and unique approach. DENSITY represents the first major addition to the analysis of trapping arrays in 20 years.Barker & White (2004) discuss how existing software such as Program MARK require that each new model’s likelihood must be programmed specifically for that model. They wishfully think that future software might allow the user to combine pieces of likelihood functions together to generate estimates. The idea is interesting, and maybe some bright young statistician can work out the specifics to implement the procedure.Choquet et al. (2004) describe MSURGE, a software package that implements the multistate capture–recapture models. The unique feature of MSURGE is that the design matrix is constructed with an interpreted language called GEMACO. Because MSURGE is limited to just multistate models, the special requirements of these likelihoods can be provided.The software and methods presented in these papers gives biologists and wildlife managers an expanding range of possibilities for data analysis. Although ease–of–use is generally getting better, it does not replace the need for understanding of the requirements and structure of the models being computed. The internet provides access to many free software packages as well as user–discussion groups to share knowledge and ideas. (A starting point for wildlife–related applications is (http://www.phidot.org).

Evidence-Based Occupational Hearing Screening I: Modeling the Effects of Real-World Noise Environments on the Likelihood of Effective Speech Communication.

PubMed

Soli, Sigfrid D; Giguère, Christian; Laroche, Chantal; Vaillancourt, Véronique; Dreschler, Wouter A; Rhebergen, Koenraad S; Harkins, Kevin; Ruckstuhl, Mark; Ramulu, Pradeep; Meyers, Lawrence S

The objectives of this study were to (1) identify essential hearing-critical job tasks for public safety and law enforcement personnel; (2) determine the locations and real-world noise environments where these tasks are performed; (3) characterize each noise environment in terms of its impact on the likelihood of effective speech communication, considering the effects of different levels of vocal effort, communication distances, and repetition; and (4) use this characterization to define an objective normative reference for evaluating the ability of individuals to perform essential hearing-critical job tasks in noisy real-world environments. Data from five occupational hearing studies performed over a 17-year period for various public safety agencies were analyzed. In each study, job task analyses by job content experts identified essential hearing-critical tasks and the real-world noise environments where these tasks are performed. These environments were visited, and calibrated recordings of each noise environment were made. The extended speech intelligibility index (ESII) was calculated for each 4-sec interval in each recording. These data, together with the estimated ESII value required for effective speech communication by individuals with normal hearing, allowed the likelihood of effective speech communication in each noise environment for different levels of vocal effort and communication distances to be determined. These likelihoods provide an objective norm-referenced and standardized means of characterizing the predicted impact of real-world noise on the ability to perform essential hearing-critical tasks. A total of 16 noise environments for law enforcement personnel and eight noise environments for corrections personnel were analyzed. Effective speech communication was essential to hearing-critical tasks performed in these environments. Average noise levels, ranged from approximately 70 to 87 dBA in law enforcement environments and 64 to 80 dBA in corrections environments. The likelihood of effective speech communication at communication distances of 0.5 and 1 m was often less than 0.50 for normal vocal effort. Likelihood values often increased to 0.80 or more when raised or loud vocal effort was used. Effective speech communication at and beyond 5 m was often unlikely, regardless of vocal effort. ESII modeling of nonstationary real-world noise environments may prove an objective means of characterizing their impact on the likelihood of effective speech communication. The normative reference provided by these measures predicts the extent to which hearing impairments that increase the ESII value required for effective speech communication also decrease the likelihood of effective speech communication. These predictions may provide an objective evidence-based link between the essential hearing-critical job task requirements of public safety and law enforcement personnel and ESII-based hearing assessment of individuals who seek to perform these jobs.

Maximum likelihood estimation of signal detection model parameters for the assessment of two-stage diagnostic strategies.

PubMed

Lirio, R B; Dondériz, I C; Pérez Abalo, M C

1992-08-01

The methodology of Receiver Operating Characteristic curves based on the signal detection model is extended to evaluate the accuracy of two-stage diagnostic strategies. A computer program is developed for the maximum likelihood estimation of parameters that characterize the sensitivity and specificity of two-stage classifiers according to this extended methodology. Its use is briefly illustrated with data collected in a two-stage screening for auditory defects.

Computing Maximum Likelihood Estimates of Loglinear Models from Marginal Sums with Special Attention to Loglinear Item Response Theory. [Project Psychometric Aspects of Item Banking No. 53.] Research Report 91-1.

ERIC Educational Resources Information Center

Kelderman, Henk

In this paper, algorithms are described for obtaining the maximum likelihood estimates of the parameters in log-linear models. Modified versions of the iterative proportional fitting and Newton-Raphson algorithms are described that work on the minimal sufficient statistics rather than on the usual counts in the full contingency table. This is…

Model selection and parameter estimation in structural dynamics using approximate Bayesian computation

NASA Astrophysics Data System (ADS)

Ben Abdessalem, Anis; Dervilis, Nikolaos; Wagg, David; Worden, Keith

2018-01-01

This paper will introduce the use of the approximate Bayesian computation (ABC) algorithm for model selection and parameter estimation in structural dynamics. ABC is a likelihood-free method typically used when the likelihood function is either intractable or cannot be approached in a closed form. To circumvent the evaluation of the likelihood function, simulation from a forward model is at the core of the ABC algorithm. The algorithm offers the possibility to use different metrics and summary statistics representative of the data to carry out Bayesian inference. The efficacy of the algorithm in structural dynamics is demonstrated through three different illustrative examples of nonlinear system identification: cubic and cubic-quintic models, the Bouc-Wen model and the Duffing oscillator. The obtained results suggest that ABC is a promising alternative to deal with model selection and parameter estimation issues, specifically for systems with complex behaviours.

Aircraft parameter estimation

NASA Technical Reports Server (NTRS)

Iliff, Kenneth W.

1987-01-01

The aircraft parameter estimation problem is used to illustrate the utility of parameter estimation, which applies to many engineering and scientific fields. Maximum likelihood estimation has been used to extract stability and control derivatives from flight data for many years. This paper presents some of the basic concepts of aircraft parameter estimation and briefly surveys the literature in the field. The maximum likelihood estimator is discussed, and the basic concepts of minimization and estimation are examined for a simple simulated aircraft example. The cost functions that are to be minimized during estimation are defined and discussed. Graphic representations of the cost functions are given to illustrate the minimization process. Finally, the basic concepts are generalized, and estimation from flight data is discussed. Some of the major conclusions for the simulated example are also developed for the analysis of flight data from the F-14, highly maneuverable aircraft technology (HiMAT), and space shuttle vehicles.

The meaning of diagnostic test results: a spreadsheet for swift data analysis.

PubMed

Maceneaney, P M; Malone, D E

2000-03-01

To design a spreadsheet program to: (a) analyse rapidly diagnostic test result data produced in local research or reported in the literature; (b) correct reported predictive values for disease prevalence in any population; (c) estimate the post-test probability of disease in individual patients. Microsoft Excel(TM)was used. Section A: a contingency (2 x 2) table was incorporated into the spreadsheet. Formulae for standard calculations [sample size, disease prevalence, sensitivity and specificity with 95% confidence intervals, predictive values and likelihood ratios (LRs)] were linked to this table. The results change automatically when the data in the true or false negative and positive cells are changed. Section B: this estimates predictive values in any population, compensating for altered disease prevalence. Sections C-F: Bayes' theorem was incorporated to generate individual post-test probabilities. The spreadsheet generates 95% confidence intervals, LRs and a table and graph of conditional probabilities once the sensitivity and specificity of the test are entered. The latter shows the expected post-test probability of disease for any pre-test probability when a test of known sensitivity and specificity is positive or negative. This spreadsheet can be used on desktop and palmtop computers. The MS Excel(TM)version can be downloaded via the Internet from the URL ftp://radiography.com/pub/Rad-data99.xls A spreadsheet is useful for contingency table data analysis and assessment of the clinical meaning of diagnostic test results. Copyright 2000 The Royal College of Radiologists.

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.

Fast and accurate estimation of the covariance between pairwise maximum likelihood distances.

PubMed

Gil, Manuel

2014-01-01

Pairwise evolutionary distances are a model-based summary statistic for a set of molecular sequences. They represent the leaf-to-leaf path lengths of the underlying phylogenetic tree. Estimates of pairwise distances with overlapping paths covary because of shared mutation events. It is desirable to take these covariance structure into account to increase precision in any process that compares or combines distances. This paper introduces a fast estimator for the covariance of two pairwise maximum likelihood distances, estimated under general Markov models. The estimator is based on a conjecture (going back to Nei & Jin, 1989) which links the covariance to path lengths. It is proven here under a simple symmetric substitution model. A simulation shows that the estimator outperforms previously published ones in terms of the mean squared error.

Fast and accurate estimation of the covariance between pairwise maximum likelihood distances

PubMed Central

2014-01-01

Pairwise evolutionary distances are a model-based summary statistic for a set of molecular sequences. They represent the leaf-to-leaf path lengths of the underlying phylogenetic tree. Estimates of pairwise distances with overlapping paths covary because of shared mutation events. It is desirable to take these covariance structure into account to increase precision in any process that compares or combines distances. This paper introduces a fast estimator for the covariance of two pairwise maximum likelihood distances, estimated under general Markov models. The estimator is based on a conjecture (going back to Nei & Jin, 1989) which links the covariance to path lengths. It is proven here under a simple symmetric substitution model. A simulation shows that the estimator outperforms previously published ones in terms of the mean squared error. PMID:25279263

Using a network-based approach and targeted maximum likelihood estimation to evaluate the effect of adding pre-exposure prophylaxis to an ongoing test-and-treat trial.

PubMed

Balzer, Laura; Staples, Patrick; Onnela, Jukka-Pekka; DeGruttola, Victor

2017-04-01

Several cluster-randomized trials are underway to investigate the implementation and effectiveness of a universal test-and-treat strategy on the HIV epidemic in sub-Saharan Africa. We consider nesting studies of pre-exposure prophylaxis within these trials. Pre-exposure prophylaxis is a general strategy where high-risk HIV- persons take antiretrovirals daily to reduce their risk of infection from exposure to HIV. We address how to target pre-exposure prophylaxis to high-risk groups and how to maximize power to detect the individual and combined effects of universal test-and-treat and pre-exposure prophylaxis strategies. We simulated 1000 trials, each consisting of 32 villages with 200 individuals per village. At baseline, we randomized the universal test-and-treat strategy. Then, after 3 years of follow-up, we considered four strategies for targeting pre-exposure prophylaxis: (1) all HIV- individuals who self-identify as high risk, (2) all HIV- individuals who are identified by their HIV+ partner (serodiscordant couples), (3) highly connected HIV- individuals, and (4) the HIV- contacts of a newly diagnosed HIV+ individual (a ring-based strategy). We explored two possible trial designs, and all villages were followed for a total of 7 years. For each village in a trial, we used a stochastic block model to generate bipartite (male-female) networks and simulated an agent-based epidemic process on these networks. We estimated the individual and combined intervention effects with a novel targeted maximum likelihood estimator, which used cross-validation to data-adaptively select from a pre-specified library the candidate estimator that maximized the efficiency of the analysis. The universal test-and-treat strategy reduced the 3-year cumulative HIV incidence by 4.0% on average. The impact of each pre-exposure prophylaxis strategy on the 4-year cumulative HIV incidence varied by the coverage of the universal test-and-treat strategy with lower coverage resulting in a larger impact of pre-exposure prophylaxis. Offering pre-exposure prophylaxis to serodiscordant couples resulted in the largest reductions in HIV incidence (2% reduction), and the ring-based strategy had little impact (0% reduction). The joint effect was larger than either individual effect with reductions in the 7-year incidence ranging from 4.5% to 8.8%. Targeted maximum likelihood estimation, data-adaptively adjusting for baseline covariates, substantially improved power over the unadjusted analysis, while maintaining nominal confidence interval coverage. Our simulation study suggests that nesting a pre-exposure prophylaxis study within an ongoing trial can lead to combined intervention effects greater than those of universal test-and-treat alone and can provide information about the efficacy of pre-exposure prophylaxis in the presence of high coverage of treatment for HIV+ persons.

Technical Note: Approximate Bayesian parameterization of a process-based tropical forest model

NASA Astrophysics Data System (ADS)

Hartig, F.; Dislich, C.; Wiegand, T.; Huth, A.

2014-02-01

Inverse parameter estimation of process-based models is a long-standing problem in many scientific disciplines. A key question for inverse parameter estimation is how to define the metric that quantifies how well model predictions fit to the data. This metric can be expressed by general cost or objective functions, but statistical inversion methods require a particular metric, the probability of observing the data given the model parameters, known as the likelihood. For technical and computational reasons, likelihoods for process-based stochastic models are usually based on general assumptions about variability in the observed data, and not on the stochasticity generated by the model. Only in recent years have new methods become available that allow the generation of likelihoods directly from stochastic simulations. Previous applications of these approximate Bayesian methods have concentrated on relatively simple models. Here, we report on the application of a simulation-based likelihood approximation for FORMIND, a parameter-rich individual-based model of tropical forest dynamics. We show that approximate Bayesian inference, based on a parametric likelihood approximation placed in a conventional Markov chain Monte Carlo (MCMC) sampler, performs well in retrieving known parameter values from virtual inventory data generated by the forest model. We analyze the results of the parameter estimation, examine its sensitivity to the choice and aggregation of model outputs and observed data (summary statistics), and demonstrate the application of this method by fitting the FORMIND model to field data from an Ecuadorian tropical forest. Finally, we discuss how this approach differs from approximate Bayesian computation (ABC), another method commonly used to generate simulation-based likelihood approximations. Our results demonstrate that simulation-based inference, which offers considerable conceptual advantages over more traditional methods for inverse parameter estimation, can be successfully applied to process-based models of high complexity. The methodology is particularly suitable for heterogeneous and complex data structures and can easily be adjusted to other model types, including most stochastic population and individual-based models. Our study therefore provides a blueprint for a fairly general approach to parameter estimation of stochastic process-based models.

Technical Note: Approximate Bayesian parameterization of a complex tropical forest model

NASA Astrophysics Data System (ADS)

Hartig, F.; Dislich, C.; Wiegand, T.; Huth, A.

2013-08-01

Inverse parameter estimation of process-based models is a long-standing problem in ecology and evolution. A key problem of inverse parameter estimation is to define a metric that quantifies how well model predictions fit to the data. Such a metric can be expressed by general cost or objective functions, but statistical inversion approaches are based on a particular metric, the probability of observing the data given the model, known as the likelihood. Deriving likelihoods for dynamic models requires making assumptions about the probability for observations to deviate from mean model predictions. For technical reasons, these assumptions are usually derived without explicit consideration of the processes in the simulation. Only in recent years have new methods become available that allow generating likelihoods directly from stochastic simulations. Previous applications of these approximate Bayesian methods have concentrated on relatively simple models. Here, we report on the application of a simulation-based likelihood approximation for FORMIND, a parameter-rich individual-based model of tropical forest dynamics. We show that approximate Bayesian inference, based on a parametric likelihood approximation placed in a conventional MCMC, performs well in retrieving known parameter values from virtual field data generated by the forest model. We analyze the results of the parameter estimation, examine the sensitivity towards the choice and aggregation of model outputs and observed data (summary statistics), and show results from using this method to fit the FORMIND model to field data from an Ecuadorian tropical forest. Finally, we discuss differences of this approach to Approximate Bayesian Computing (ABC), another commonly used method to generate simulation-based likelihood approximations. Our results demonstrate that simulation-based inference, which offers considerable conceptual advantages over more traditional methods for inverse parameter estimation, can successfully be applied to process-based models of high complexity. The methodology is particularly suited to heterogeneous and complex data structures and can easily be adjusted to other model types, including most stochastic population and individual-based models. Our study therefore provides a blueprint for a fairly general approach to parameter estimation of stochastic process-based models in ecology and evolution.

The 1993 Mississippi river flood: A one hundred or a one thousand year event?

USGS Publications Warehouse

Malamud, B.D.; Turcotte, D.L.; Barton, C.C.

1996-01-01

Power-law (fractal) extreme-value statistics are applicable to many natural phenomena under a wide variety of circumstances. Data from a hydrologic station in Keokuk, Iowa, shows the great flood of the Mississippi River in 1993 has a recurrence interval on the order of 100 years using power-law statistics applied to partial-duration flood series and on the order of 1,000 years using a log-Pearson type 3 (LP3) distribution applied to annual series. The LP3 analysis is the federally adopted probability distribution for flood-frequency estimation of extreme events. We suggest that power-law statistics are preferable to LP3 analysis. As a further test of the power-law approach we consider paleoflood data from the Colorado River. We compare power-law and LP3 extrapolations of historical data with these paleo-floods. The results are remarkably similar to those obtained for the Mississippi River: Recurrence intervals from power-law statistics applied to Lees Ferry discharge data are generally consistent with inferred 100- and 1,000-year paleofloods, whereas LP3 analysis gives recurrence intervals that are orders of magnitude longer. For both the Keokuk and Lees Ferry gauges, the use of an annual series introduces an artificial curvature in log-log space that leads to an underestimate of severe floods. Power-law statistics are predicting much shorter recurrence intervals than the federally adopted LP3 statistics. We suggest that if power-law behavior is applicable, then the likelihood of severe floods is much higher. More conservative dam designs and land-use restrictions Nay be required.

Semiparametric regression analysis of failure time data with dependent interval censoring.

PubMed

Chen, Chyong-Mei; Shen, Pao-Sheng

2017-09-20

Interval-censored failure-time data arise when subjects are examined or observed periodically such that the failure time of interest is not examined exactly but only known to be bracketed between two adjacent observation times. The commonly used approaches assume that the examination times and the failure time are independent or conditionally independent given covariates. In many practical applications, patients who are already in poor health or have a weak immune system before treatment usually tend to visit physicians more often after treatment than those with better health or immune system. In this situation, the visiting rate is positively correlated with the risk of failure due to the health status, which results in dependent interval-censored data. While some measurable factors affecting health status such as age, gender, and physical symptom can be included in the covariates, some health-related latent variables cannot be observed or measured. To deal with dependent interval censoring involving unobserved latent variable, we characterize the visiting/examination process as recurrent event process and propose a joint frailty model to account for the association of the failure time and visiting process. A shared gamma frailty is incorporated into the Cox model and proportional intensity model for the failure time and visiting process, respectively, in a multiplicative way. We propose a semiparametric maximum likelihood approach for estimating model parameters and show the asymptotic properties, including consistency and weak convergence. Extensive simulation studies are conducted and a data set of bladder cancer is analyzed for illustrative purposes. Copyright © 2017 John Wiley & Sons, Ltd. Copyright © 2017 John Wiley & Sons, Ltd.

Psychometric Properties of IRT Proficiency Estimates

ERIC Educational Resources Information Center

Kolen, Michael J.; Tong, Ye

2010-01-01

Psychometric properties of item response theory proficiency estimates are considered in this paper. Proficiency estimators based on summed scores and pattern scores include non-Bayes maximum likelihood and test characteristic curve estimators and Bayesian estimators. The psychometric properties investigated include reliability, conditional…

Quantum-state reconstruction by maximizing likelihood and entropy.

PubMed

Teo, Yong Siah; Zhu, Huangjun; Englert, Berthold-Georg; Řeháček, Jaroslav; Hradil, Zdeněk

2011-07-08

Quantum-state reconstruction on a finite number of copies of a quantum system with informationally incomplete measurements, as a rule, does not yield a unique result. We derive a reconstruction scheme where both the likelihood and the von Neumann entropy functionals are maximized in order to systematically select the most-likely estimator with the largest entropy, that is, the least-bias estimator, consistent with a given set of measurement data. This is equivalent to the joint consideration of our partial knowledge and ignorance about the ensemble to reconstruct its identity. An interesting structure of such estimators will also be explored.

Extreme geomagnetic storms: Probabilistic forecasts and their uncertainties

USGS Publications Warehouse

Riley, Pete; Love, Jeffrey J.

2017-01-01

Extreme space weather events are low-frequency, high-risk phenomena. Estimating their rates of occurrence, as well as their associated uncertainties, is difficult. In this study, we derive statistical estimates and uncertainties for the occurrence rate of an extreme geomagnetic storm on the scale of the Carrington event (or worse) occurring within the next decade. We model the distribution of events as either a power law or lognormal distribution and use (1) Kolmogorov-Smirnov statistic to estimate goodness of fit, (2) bootstrapping to quantify the uncertainty in the estimates, and (3) likelihood ratio tests to assess whether one distribution is preferred over another. Our best estimate for the probability of another extreme geomagnetic event comparable to the Carrington event occurring within the next 10 years is 10.3% 95%  confidence interval (CI) [0.9,18.7] for a power law distribution but only 3.0% 95% CI [0.6,9.0] for a lognormal distribution. However, our results depend crucially on (1) how we define an extreme event, (2) the statistical model used to describe how the events are distributed in intensity, (3) the techniques used to infer the model parameters, and (4) the data and duration used for the analysis. We test a major assumption that the data represent time stationary processes and discuss the implications. If the current trends persist, suggesting that we are entering a period of lower activity, our forecasts may represent upper limits rather than best estimates.

Likelihoods for fixed rank nomination networks

PubMed Central

HOFF, PETER; FOSDICK, BAILEY; VOLFOVSKY, ALEX; STOVEL, KATHERINE

2014-01-01

Many studies that gather social network data use survey methods that lead to censored, missing, or otherwise incomplete information. For example, the popular fixed rank nomination (FRN) scheme, often used in studies of schools and businesses, asks study participants to nominate and rank at most a small number of contacts or friends, leaving the existence of other relations uncertain. However, most statistical models are formulated in terms of completely observed binary networks. Statistical analyses of FRN data with such models ignore the censored and ranked nature of the data and could potentially result in misleading statistical inference. To investigate this possibility, we compare Bayesian parameter estimates obtained from a likelihood for complete binary networks with those obtained from likelihoods that are derived from the FRN scheme, and therefore accommodate the ranked and censored nature of the data. We show analytically and via simulation that the binary likelihood can provide misleading inference, particularly for certain model parameters that relate network ties to characteristics of individuals and pairs of individuals. We also compare these different likelihoods in a data analysis of several adolescent social networks. For some of these networks, the parameter estimates from the binary and FRN likelihoods lead to different conclusions, indicating the importance of analyzing FRN data with a method that accounts for the FRN survey design. PMID:25110586

Modeling motor vehicle crashes using Poisson-gamma models: examining the effects of low sample mean values and small sample size on the estimation of the fixed dispersion parameter.

PubMed

Lord, Dominique

2006-07-01

There has been considerable research conducted on the development of statistical models for predicting crashes on highway facilities. Despite numerous advancements made for improving the estimation tools of statistical models, the most common probabilistic structure used for modeling motor vehicle crashes remains the traditional Poisson and Poisson-gamma (or Negative Binomial) distribution; when crash data exhibit over-dispersion, the Poisson-gamma model is usually the model of choice most favored by transportation safety modelers. Crash data collected for safety studies often have the unusual attributes of being characterized by low sample mean values. Studies have shown that the goodness-of-fit of statistical models produced from such datasets can be significantly affected. This issue has been defined as the "low mean problem" (LMP). Despite recent developments on methods to circumvent the LMP and test the goodness-of-fit of models developed using such datasets, no work has so far examined how the LMP affects the fixed dispersion parameter of Poisson-gamma models used for modeling motor vehicle crashes. The dispersion parameter plays an important role in many types of safety studies and should, therefore, be reliably estimated. The primary objective of this research project was to verify whether the LMP affects the estimation of the dispersion parameter and, if it is, to determine the magnitude of the problem. The secondary objective consisted of determining the effects of an unreliably estimated dispersion parameter on common analyses performed in highway safety studies. To accomplish the objectives of the study, a series of Poisson-gamma distributions were simulated using different values describing the mean, the dispersion parameter, and the sample size. Three estimators commonly used by transportation safety modelers for estimating the dispersion parameter of Poisson-gamma models were evaluated: the method of moments, the weighted regression, and the maximum likelihood method. In an attempt to complement the outcome of the simulation study, Poisson-gamma models were fitted to crash data collected in Toronto, Ont. characterized by a low sample mean and small sample size. The study shows that a low sample mean combined with a small sample size can seriously affect the estimation of the dispersion parameter, no matter which estimator is used within the estimation process. The probability the dispersion parameter becomes unreliably estimated increases significantly as the sample mean and sample size decrease. Consequently, the results show that an unreliably estimated dispersion parameter can significantly undermine empirical Bayes (EB) estimates as well as the estimation of confidence intervals for the gamma mean and predicted response. The paper ends with recommendations about minimizing the likelihood of producing Poisson-gamma models with an unreliable dispersion parameter for modeling motor vehicle crashes.

Lateral stability and control derivatives of a jet fighter airplane extracted from flight test data by utilizing maximum likelihood estimation

NASA Technical Reports Server (NTRS)

Parrish, R. V.; Steinmetz, G. G.

1972-01-01

A method of parameter extraction for stability and control derivatives of aircraft from flight test data, implementing maximum likelihood estimation, has been developed and successfully applied to actual lateral flight test data from a modern sophisticated jet fighter. This application demonstrates the important role played by the analyst in combining engineering judgment and estimator statistics to yield meaningful results. During the analysis, the problems of uniqueness of the extracted set of parameters and of longitudinal coupling effects were encountered and resolved. The results for all flight runs are presented in tabular form and as time history comparisons between the estimated states and the actual flight test data.

Effect of sampling rate and record length on the determination of stability and control derivatives

NASA Technical Reports Server (NTRS)

Brenner, M. J.; Iliff, K. W.; Whitman, R. K.

1978-01-01

Flight data from five aircraft were used to assess the effects of sampling rate and record length reductions on estimates of stability and control derivatives produced by a maximum likelihood estimation method. Derivatives could be extracted from flight data with the maximum likelihood estimation method even if there were considerable reductions in sampling rate and/or record length. Small amplitude pulse maneuvers showed greater degradation of the derivative maneuvers than large amplitude pulse maneuvers when these reductions were made. Reducing the sampling rate was found to be more desirable than reducing the record length as a method of lessening the total computation time required without greatly degrading the quantity of the estimates.

Characterization, parameter estimation, and aircraft response statistics of atmospheric turbulence

NASA Technical Reports Server (NTRS)

Mark, W. D.

1981-01-01

A nonGaussian three component model of atmospheric turbulence is postulated that accounts for readily observable features of turbulence velocity records, their autocorrelation functions, and their spectra. Methods for computing probability density functions and mean exceedance rates of a generic aircraft response variable are developed using nonGaussian turbulence characterizations readily extracted from velocity recordings. A maximum likelihood method is developed for optimal estimation of the integral scale and intensity of records possessing von Karman transverse of longitudinal spectra. Formulas for the variances of such parameter estimates are developed. The maximum likelihood and least-square approaches are combined to yield a method for estimating the autocorrelation function parameters of a two component model for turbulence.

Objectively combining AR5 instrumental period and paleoclimate climate sensitivity evidence

NASA Astrophysics Data System (ADS)

Lewis, Nicholas; Grünwald, Peter

2018-03-01

Combining instrumental period evidence regarding equilibrium climate sensitivity with largely independent paleoclimate proxy evidence should enable a more constrained sensitivity estimate to be obtained. Previous, subjective Bayesian approaches involved selection of a prior probability distribution reflecting the investigators' beliefs about climate sensitivity. Here a recently developed approach employing two different statistical methods—objective Bayesian and frequentist likelihood-ratio—is used to combine instrumental period and paleoclimate evidence based on data presented and assessments made in the IPCC Fifth Assessment Report. Probabilistic estimates from each source of evidence are represented by posterior probability density functions (PDFs) of physically-appropriate form that can be uniquely factored into a likelihood function and a noninformative prior distribution. The three-parameter form is shown accurately to fit a wide range of estimated climate sensitivity PDFs. The likelihood functions relating to the probabilistic estimates from the two sources are multiplicatively combined and a prior is derived that is noninformative for inference from the combined evidence. A posterior PDF that incorporates the evidence from both sources is produced using a single-step approach, which avoids the order-dependency that would arise if Bayesian updating were used. Results are compared with an alternative approach using the frequentist signed root likelihood ratio method. Results from these two methods are effectively identical, and provide a 5-95% range for climate sensitivity of 1.1-4.05 K (median 1.87 K).

Maximal likelihood correspondence estimation for face recognition across pose.

PubMed

Li, Shaoxin; Liu, Xin; Chai, Xiujuan; Zhang, Haihong; Lao, Shihong; Shan, Shiguang

2014-10-01

Due to the misalignment of image features, the performance of many conventional face recognition methods degrades considerably in across pose scenario. To address this problem, many image matching-based methods are proposed to estimate semantic correspondence between faces in different poses. In this paper, we aim to solve two critical problems in previous image matching-based correspondence learning methods: 1) fail to fully exploit face specific structure information in correspondence estimation and 2) fail to learn personalized correspondence for each probe image. To this end, we first build a model, termed as morphable displacement field (MDF), to encode face specific structure information of semantic correspondence from a set of real samples of correspondences calculated from 3D face models. Then, we propose a maximal likelihood correspondence estimation (MLCE) method to learn personalized correspondence based on maximal likelihood frontal face assumption. After obtaining the semantic correspondence encoded in the learned displacement, we can synthesize virtual frontal images of the profile faces for subsequent recognition. Using linear discriminant analysis method with pixel-intensity features, state-of-the-art performance is achieved on three multipose benchmarks, i.e., CMU-PIE, FERET, and MultiPIE databases. Owe to the rational MDF regularization and the usage of novel maximal likelihood objective, the proposed MLCE method can reliably learn correspondence between faces in different poses even in complex wild environment, i.e., labeled face in the wild database.

Knowledge and risk perception of late effects among childhood cancer survivors and parents before and after visiting a childhood cancer survivor clinic.

PubMed

Cherven, Brooke; Mertens, Ann; Meacham, Lillian R; Williamson, Rebecca; Boring, Cathy; Wasilewski-Masker, Karen

2014-01-01

Survivors of childhood cancer are at risk for a variety of treatment-related late effects and require lifelong individualized surveillance for early detection of late effects. This study assessed knowledge and perceptions of late effects risk before and after a survivor clinic visit. Young adult survivors (≥ 16 years) and parents of child survivors (< 16 years) were recruited prior to initial visit to a cancer survivor program. Sixty-five participants completed a baseline survey and 50 completed both a baseline and follow-up survey. Participants were found to have a low perceived likelihood of developing a late effect of cancer therapy and many incorrect perceptions of risk for individual late effects. Low knowledge before clinic (odds ratio = 9.6; 95% confidence interval, 1.7-92.8; P = .02) and low perceived likelihood of developing a late effect (odds ratio = 18.7; 95% confidence interval, 2.7-242.3; P = .01) were found to predict low knowledge of late effect risk at follow-up. This suggests that perceived likelihood of developing a late effect is an important factor in the individuals' ability to learn about their risk and should be addressed before initiation of education. © 2014 by Association of Pediatric Hematology/Oncology Nurses.

Variability of space climate and its extremes with successive solar cycles

NASA Astrophysics Data System (ADS)

Chapman, Sandra; Hush, Phillip; Tindale, Elisabeth; Dunlop, Malcolm; Watkins, Nicholas

2016-04-01

Auroral geomagnetic indices coupled with in situ solar wind monitors provide a comprehensive data set, spanning several solar cycles. Space climate can be considered as the distribution of space weather. We can then characterize these observations in terms of changing space climate by quantifying how the statistical properties of ensembles of these observed variables vary between different phases of the solar cycle. We first consider the AE index burst distribution. Bursts are constructed by thresholding the AE time series; the size of a burst is the sum of the excess in the time series for each time interval over which the threshold is exceeded. The distribution of burst sizes is two component with a crossover in behaviour at thresholds ≈ 1000 nT. Above this threshold, we find[1] a range over which the mean burst size is almost constant with threshold for both solar maxima and minima. The burst size distribution of the largest events has a functional form which is exponential. The relative likelihood of these large events varies from one solar maximum and minimum to the next. If the relative overall activity of a solar maximum/minimum can be estimated, these results then constrain the likelihood of extreme events of a given size for that solar maximum/minimum. We next develop and apply a methodology to quantify how the full distribution of geomagnetic indices and upstream solar wind observables are changing between and across different solar cycles. This methodology[2] estimates how different quantiles of the distribution, or equivalently, how the return times of events of a given size, are changing. [1] Hush, P., S. C. Chapman, M. W. Dunlop, and N. W. Watkins (2015), Robust statistical properties of the size of large burst events in AE, Geophys. Res. Lett.,42 doi:10.1002/2015GL066277 [2] Chapman, S. C., D. A. Stainforth, N. W. Watkins, (2013) On estimating long term local climate trends , Phil. Trans. Royal Soc., A,371 20120287 DOI:10.1098/rsta.2012.0287

Impact of the Adalimumab Patient Support Program on Clinical Outcomes in Ankylosing Spondylitis: Results from the COMPANION Study.

PubMed

Bessette, Louis; Lebovic, Gerald; Millson, Brad; Charland, Katia; Donepudi, Krishna; Gaetano, Tania; Remple, Valencia; Latour, Martin G; Gazel, Sandra; Laliberté, Marie-Claude; Thorne, Carter

2018-06-01

Adalimumab (ADA) is a tumor necrosis factor (TNF)-alpha inhibitor indicated for the treatment of inflammatory autoimmune diseases, including ankylosing spondylitis (AS). Patients receiving ADA in Canada are eligible to enroll in the AbbVie Care™ patient support program (AC-PSP), which provides personalized services, including care coach calls (CCCs). We estimated the likelihood of controlled disease in a cohort of AS patients treated with ADA enrolled in the AC-PSP and who received CCCs versus those who did not. A longitudinal analysis using de-identified aggregate-level data collected through the AC-PSP was performed. A probabilistic matching algorithm was used to link patient-level records from the AC-PSP database to records from the QuintilesIMS longitudinal prescription transactions database. Patients were indexed on the date of their first prescription of ADA between January 2010 and October 2015. The AC-PSP database included patient assessments of the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), a measure of disease activity. Eligible patients had a baseline BASDAI assessment performed between 90 days before and 30 days after the index date, and a follow-up BASDAI assessment 6-18 months later. Poisson regression was used to estimate the adjusted relative risk (RR) of controlled disease (BASDAI < 4) at the time of follow-up, comparing patients who received CCCs with those who did not. In total 249 AS patients met eligibility criteria, and 123 (49%) received CCCs. Of the 249 patients, 184 (74%) had controlled disease (BASDAI < 4) at follow-up assessment, 98 (80%) in the CCC group and 86 (68%) in the no CCC group. Multivariable regression analysis demonstrated a 23% increased likelihood of controlled disease in patients who received CCCs relative to those who did not (RR = 1.23; 95% confidence interval, 1.06-1.42; p = 0.0055). AS patients receiving tailored services through the AC-PSP in the form of CCCs have an increased likelihood of controlled disease within 6-18 months. AbbVie.

Overturning conclusions of Lévy flight movement patterns by fishing boats and foraging animals.

PubMed

Edwards, Andrew M

2011-06-01

A surprisingly diverse variety of foragers have previously been concluded to exhibit movement patterns known as Lévy flights, a special type of random walk. These foragers range in size from microzooplankton in experiments to fishermen in the Pacific Ocean and the North Sea. The Lévy flight conclusion implies that all the foragers have similar scale-free movement patterns that can be described by a single dimensionless parameter, the exponent micro of a power-law (Pareto) distribution. However, the previous conclusions have been made using methods that have since been shown to be problematic: inaccurate techniques were used to estimate micro, and the power-law distribution was usually assumed to hold without testing any alternative hypotheses. Therefore, I address the open question of whether the previous data still support the Lévy flight hypothesis, and thus determine whether Lévy flights really are so ubiquitous in ecology. I present a comprehensive reanalysis of 17 data sets from seven previous studies for which Lévy flight behavior had been concluded, covering marine, terrestrial, and experimental systems from four continents. I use the modern likelihood and Akaike weights approach to test whether simple alternative models are more supported by the data than Lévy flights. The previously estimated values of the power-law exponent micro do not match those calculated here using the accurate likelihood approach, and almost all of them lie outside of the likelihood-based 95% confidence intervals. Furthermore, the original power-law Lévy flight model is overwhelmingly rejected for 16 out of the 17 data sets when tested against three other simple models. For one data set, the data are consistent with coming from a bounded power-law distribution (a truncated Lévy flight). For three other data sets, an exponential distribution corresponding to a simple Poisson process is suitable. Thus, Lévy flight movement patterns are not the common phenomena that was once thought, and are not suitable for use as ecosystem indicators for fisheries management, as has been proposed.

Social support modifies association between forward bending of the trunk and low-back pain: Cross-sectional field study of blue-collar workers.

PubMed

Villumsen, Morten; Holtermann, Andreas; Samani, Afshin; Madeleine, Pascal; Jørgensen, Marie Birk

2016-03-01

This study aimed to investigate the association between forward bending of the trunk and low-back pain intensity (LBPi) among blue-collar workers in Denmark as well as whether the level of social support modifies the association. In total, 457 workers were included in the study. The forward bending of ≥ 30° was computed from accelerometer recordings for several consecutive days during work, categorized into long (highest tertile) and short-moderate (remaining tertiles) duration. LBPi was measured on a 0-10 scale and categorized into low (≤ 5) and high (>5) pain. Self-reported social support was categorized into low, moderate, and high levels. Multi-adjusted logistic regressions estimated the association between forward bending and LBPi and the effect modification by social support. Forward bending and LBPi were not significantly associated but modified by social support. Workers with low social support and long duration of forward bending had higher likelihood of high LBPi [odds ratio (OR) 2.97, 95% confidence interval (95% CI) 1.11-7.95] compared to workers with high social support and long duration of forward bending. Among workers with low social support, workers with long duration of forward bending had higher likelihood of high LBPi (OR 3.28, 95% CI 0.99-10.90) compared to workers with short-moderate duration of forward bending. Among workers with high social support, workers with long duration of forward bending had reduced likelihood of high LBPi (OR 0.39, 95% CI 0.16-0.95) compared to workers with short-moderate duration of forward bending. Social support modifies the association between objectively measured forward bending and LBPi among blue-collar workers.

Restricted maximum likelihood estimation of genetic principal components and smoothed covariance matrices

PubMed Central

Meyer, Karin; Kirkpatrick, Mark

2005-01-01

Principal component analysis is a widely used 'dimension reduction' technique, albeit generally at a phenotypic level. It is shown that we can estimate genetic principal components directly through a simple reparameterisation of the usual linear, mixed model. This is applicable to any analysis fitting multiple, correlated genetic effects, whether effects for individual traits or sets of random regression coefficients to model trajectories. Depending on the magnitude of genetic correlation, a subset of the principal component generally suffices to capture the bulk of genetic variation. Corresponding estimates of genetic covariance matrices are more parsimonious, have reduced rank and are smoothed, with the number of parameters required to model the dispersion structure reduced from k(k + 1)/2 to m(2k - m + 1)/2 for k effects and m principal components. Estimation of these parameters, the largest eigenvalues and pertaining eigenvectors of the genetic covariance matrix, via restricted maximum likelihood using derivatives of the likelihood, is described. It is shown that reduced rank estimation can reduce computational requirements of multivariate analyses substantially. An application to the analysis of eight traits recorded via live ultrasound scanning of beef cattle is given. PMID:15588566

A simulation study on Bayesian Ridge regression models for several collinearity levels

NASA Astrophysics Data System (ADS)

Efendi, Achmad; Effrihan

2017-12-01

When analyzing data with multiple regression model if there are collinearities, then one or several predictor variables are usually omitted from the model. However, there sometimes some reasons, for instance medical or economic reasons, the predictors are all important and should be included in the model. Ridge regression model is not uncommon in some researches to use to cope with collinearity. Through this modeling, weights for predictor variables are used for estimating parameters. The next estimation process could follow the concept of likelihood. Furthermore, for the estimation nowadays the Bayesian version could be an alternative. This estimation method does not match likelihood one in terms of popularity due to some difficulties; computation and so forth. Nevertheless, with the growing improvement of computational methodology recently, this caveat should not at the moment become a problem. This paper discusses about simulation process for evaluating the characteristic of Bayesian Ridge regression parameter estimates. There are several simulation settings based on variety of collinearity levels and sample sizes. The results show that Bayesian method gives better performance for relatively small sample sizes, and for other settings the method does perform relatively similar to the likelihood method.

Diagnostic accuracy of the Amsler grid and the preferential hyperacuity perimetry in the screening of patients with age-related macular degeneration: systematic review and meta-analysis.

PubMed

Faes, L; Bodmer, N S; Bachmann, L M; Thiel, M A; Schmid, M K

2014-07-01

To clarify the screening potential of the Amsler grid and preferential hyperacuity perimetry (PHP) in detecting or ruling out wet age-related macular degeneration (AMD). Medline, Scopus and Web of Science (by citation of reference) were searched. Checking of reference lists of review articles and of included articles complemented electronic searches. Papers were selected, assessed, and extracted in duplicate. Systematic review and meta-analysis. Twelve included studies enrolled 903 patients and allowed constructing 27 two-by-two tables. Twelve tables reported on the Amsler grid and its modifications, twelve tables reported on the PHP, one table assessed the MCPT and two tables assessed the M-charts. All but two studies had a case-control design. The pooled sensitivity of studies assessing the Amsler grid was 0.78 (95% confidence intervals; 0.64-0.87), and the pooled specificity was 0.97 (95% confidence intervals; 0.91-0.99). The corresponding positive and negative likelihood ratios were 23.1 (95% confidence intervals; 8.4-64.0) and 0.23 (95% confidence intervals; 0.14-0.39), respectively. The pooled sensitivity of studies assessing the PHP was 0.85 (95% confidence intervals; 0.80-0.89), and specificity was 0.87 (95% confidence intervals; 0.82-0.91). The corresponding positive and negative likelihood ratios were 6.7 (95% confidence intervals; 4.6-9.8) and 0.17 (95% confidence intervals; 0.13-0.23). No pooling was possible for MCPT and M-charts. Results from small preliminary studies show promising test performance characteristics both for the Amsler grid and PHP to rule out wet AMD in the screening setting. To what extent these findings can be transferred to a real clinic practice still needs to be established.

A practical method to test the validity of the standard Gumbel distribution in logit-based multinomial choice models of travel behavior

DOE PAGES

Ye, Xin; Garikapati, Venu M.; You, Daehyun; ...

2017-11-08

Most multinomial choice models (e.g., the multinomial logit model) adopted in practice assume an extreme-value Gumbel distribution for the random components (error terms) of utility functions. This distributional assumption offers a closed-form likelihood expression when the utility maximization principle is applied to model choice behaviors. As a result, model coefficients can be easily estimated using the standard maximum likelihood estimation method. However, maximum likelihood estimators are consistent and efficient only if distributional assumptions on the random error terms are valid. It is therefore critical to test the validity of underlying distributional assumptions on the error terms that form the basismore » of parameter estimation and policy evaluation. In this paper, a practical yet statistically rigorous method is proposed to test the validity of the distributional assumption on the random components of utility functions in both the multinomial logit (MNL) model and multiple discrete-continuous extreme value (MDCEV) model. Based on a semi-nonparametric approach, a closed-form likelihood function that nests the MNL or MDCEV model being tested is derived. The proposed method allows traditional likelihood ratio tests to be used to test violations of the standard Gumbel distribution assumption. Simulation experiments are conducted to demonstrate that the proposed test yields acceptable Type-I and Type-II error probabilities at commonly available sample sizes. The test is then applied to three real-world discrete and discrete-continuous choice models. For all three models, the proposed test rejects the validity of the standard Gumbel distribution in most utility functions, calling for the development of robust choice models that overcome adverse effects of violations of distributional assumptions on the error terms in random utility functions.« less

A practical method to test the validity of the standard Gumbel distribution in logit-based multinomial choice models of travel behavior

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

Ye, Xin; Garikapati, Venu M.; You, Daehyun

Most multinomial choice models (e.g., the multinomial logit model) adopted in practice assume an extreme-value Gumbel distribution for the random components (error terms) of utility functions. This distributional assumption offers a closed-form likelihood expression when the utility maximization principle is applied to model choice behaviors. As a result, model coefficients can be easily estimated using the standard maximum likelihood estimation method. However, maximum likelihood estimators are consistent and efficient only if distributional assumptions on the random error terms are valid. It is therefore critical to test the validity of underlying distributional assumptions on the error terms that form the basismore » of parameter estimation and policy evaluation. In this paper, a practical yet statistically rigorous method is proposed to test the validity of the distributional assumption on the random components of utility functions in both the multinomial logit (MNL) model and multiple discrete-continuous extreme value (MDCEV) model. Based on a semi-nonparametric approach, a closed-form likelihood function that nests the MNL or MDCEV model being tested is derived. The proposed method allows traditional likelihood ratio tests to be used to test violations of the standard Gumbel distribution assumption. Simulation experiments are conducted to demonstrate that the proposed test yields acceptable Type-I and Type-II error probabilities at commonly available sample sizes. The test is then applied to three real-world discrete and discrete-continuous choice models. For all three models, the proposed test rejects the validity of the standard Gumbel distribution in most utility functions, calling for the development of robust choice models that overcome adverse effects of violations of distributional assumptions on the error terms in random utility functions.« less

Planck intermediate results. XVI. Profile likelihoods for cosmological parameters

NASA Astrophysics Data System (ADS)

Planck Collaboration; Ade, P. A. R.; Aghanim, N.; Arnaud, M.; Ashdown, M.; Aumont, J.; Baccigalupi, C.; Banday, A. J.; Barreiro, R. B.; Bartlett, J. G.; Battaner, E.; Benabed, K.; Benoit-Lévy, A.; Bernard, J.-P.; Bersanelli, M.; Bielewicz, P.; Bobin, J.; Bonaldi, A.; Bond, J. R.; Bouchet, F. R.; Burigana, C.; Cardoso, J.-F.; Catalano, A.; Chamballu, A.; Chiang, H. C.; Christensen, P. R.; Clements, D. L.; Colombi, S.; Colombo, L. P. L.; Couchot, F.; Cuttaia, F.; Danese, L.; Davies, R. D.; Davis, R. J.; de Bernardis, P.; de Rosa, A.; de Zotti, G.; Delabrouille, J.; Dickinson, C.; Diego, J. M.; Dole, H.; Donzelli, S.; Doré, O.; Douspis, M.; Dupac, X.; Enßlin, T. A.; Eriksen, H. K.; Finelli, F.; Forni, O.; Frailis, M.; Franceschi, E.; Galeotta, S.; Galli, S.; Ganga, K.; Giard, M.; Giraud-Héraud, Y.; González-Nuevo, J.; Górski, K. M.; Gregorio, A.; Gruppuso, A.; Hansen, F. K.; Harrison, D. L.; Henrot-Versillé, S.; Hernández-Monteagudo, C.; Herranz, D.; Hildebrandt, S. R.; Hivon, E.; Hobson, M.; Holmes, W. A.; Hornstrup, A.; Hovest, W.; Huffenberger, K. M.; Jaffe, A. H.; Jaffe, T. R.; Jones, W. C.; Juvela, M.; Keihänen, E.; Keskitalo, R.; Kisner, T. S.; Kneissl, R.; Knoche, J.; Knox, L.; Kunz, M.; Kurki-Suonio, H.; Lagache, G.; Lähteenmäki, A.; Lamarre, J.-M.; Lasenby, A.; Lawrence, C. R.; Leonardi, R.; Liddle, A.; Liguori, M.; Lilje, P. B.; Linden-Vørnle, M.; López-Caniego, M.; Lubin, P. M.; Macías-Pérez, J. F.; Maffei, B.; Maino, D.; Mandolesi, N.; Maris, M.; Martin, P. G.; Martínez-González, E.; Masi, S.; Massardi, M.; Matarrese, S.; Mazzotta, P.; Melchiorri, A.; Mendes, L.; Mennella, A.; Migliaccio, M.; Mitra, S.; Miville-Deschênes, M.-A.; Moneti, A.; Montier, L.; Morgante, G.; Munshi, D.; Murphy, J. A.; Naselsky, P.; Nati, F.; Natoli, P.; Noviello, F.; Novikov, D.; Novikov, I.; Oxborrow, C. A.; Pagano, L.; Pajot, F.; Paoletti, D.; Pasian, F.; Perdereau, O.; Perotto, L.; Perrotta, F.; Pettorino, V.; Piacentini, F.; Piat, M.; Pierpaoli, E.; Pietrobon, D.; Plaszczynski∗, S.; Pointecouteau, E.; Polenta, G.; Popa, L.; Pratt, G. W.; Puget, J.-L.; Rachen, J. P.; Rebolo, R.; Reinecke, M.; Remazeilles, M.; Renault, C.; Ricciardi, S.; Riller, T.; Ristorcelli, I.; Rocha, G.; Rosset, C.; Roudier, G.; Rouillé d'Orfeuil, B.; Rubiño-Martín, J. A.; Rusholme, B.; Sandri, M.; Savelainen, M.; Savini, G.; Spencer, L. D.; Spinelli, M.; Starck, J.-L.; Sureau, F.; Sutton, D.; Suur-Uski, A.-S.; Sygnet, J.-F.; Tauber, J. A.; Terenzi, L.; Toffolatti, L.; Tomasi, M.; Tristram, M.; Tucci, M.; Umana, G.; Valenziano, L.; Valiviita, J.; Van Tent, B.; Vielva, P.; Villa, F.; Wade, L. A.; Wandelt, B. D.; White, M.; Yvon, D.; Zacchei, A.; Zonca, A.

2014-06-01

We explore the 2013 Planck likelihood function with a high-precision multi-dimensional minimizer (Minuit). This allows a refinement of the ΛCDM best-fit solution with respect to previously-released results, and the construction of frequentist confidence intervals using profile likelihoods. The agreement with the cosmological results from the Bayesian framework is excellent, demonstrating the robustness of the Planck results to the statistical methodology. We investigate the inclusion of neutrino masses, where more significant differences may appear due to the non-Gaussian nature of the posterior mass distribution. By applying the Feldman-Cousins prescription, we again obtain results very similar to those of the Bayesian methodology. However, the profile-likelihood analysis of the cosmic microwave background (CMB) combination (Planck+WP+highL) reveals a minimum well within the unphysical negative-mass region. We show that inclusion of the Planck CMB-lensing information regularizes this issue, and provide a robust frequentist upper limit ∑ mν ≤ 0.26 eV (95% confidence) from the CMB+lensing+BAO data combination.

Maximum likelihood method for estimating airplane stability and control parameters from flight data in frequency domain

NASA Technical Reports Server (NTRS)

Klein, V.

1980-01-01

A frequency domain maximum likelihood method is developed for the estimation of airplane stability and control parameters from measured data. The model of an airplane is represented by a discrete-type steady state Kalman filter with time variables replaced by their Fourier series expansions. The likelihood function of innovations is formulated, and by its maximization with respect to unknown parameters the estimation algorithm is obtained. This algorithm is then simplified to the output error estimation method with the data in the form of transformed time histories, frequency response curves, or spectral and cross-spectral densities. The development is followed by a discussion on the equivalence of the cost function in the time and frequency domains, and on advantages and disadvantages of the frequency domain approach. The algorithm developed is applied in four examples to the estimation of longitudinal parameters of a general aviation airplane using computer generated and measured data in turbulent and still air. The cost functions in the time and frequency domains are shown to be equivalent; therefore, both approaches are complementary and not contradictory. Despite some computational advantages of parameter estimation in the frequency domain, this approach is limited to linear equations of motion with constant coefficients.

Exponential series approaches for nonparametric graphical models

NASA Astrophysics Data System (ADS)

Janofsky, Eric

Markov Random Fields (MRFs) or undirected graphical models are parsimonious representations of joint probability distributions. This thesis studies high-dimensional, continuous-valued pairwise Markov Random Fields. We are particularly interested in approximating pairwise densities whose logarithm belongs to a Sobolev space. For this problem we propose the method of exponential series which approximates the log density by a finite-dimensional exponential family with the number of sufficient statistics increasing with the sample size. We consider two approaches to estimating these models. The first is regularized maximum likelihood. This involves optimizing the sum of the log-likelihood of the data and a sparsity-inducing regularizer. We then propose a variational approximation to the likelihood based on tree-reweighted, nonparametric message passing. This approximation allows for upper bounds on risk estimates, leverages parallelization and is scalable to densities on hundreds of nodes. We show how the regularized variational MLE may be estimated using a proximal gradient algorithm. We then consider estimation using regularized score matching. This approach uses an alternative scoring rule to the log-likelihood, which obviates the need to compute the normalizing constant of the distribution. For general continuous-valued exponential families, we provide parameter and edge consistency results. As a special case we detail a new approach to sparse precision matrix estimation which has statistical performance competitive with the graphical lasso and computational performance competitive with the state-of-the-art glasso algorithm. We then describe results for model selection in the nonparametric pairwise model using exponential series. The regularized score matching problem is shown to be a convex program; we provide scalable algorithms based on consensus alternating direction method of multipliers (ADMM) and coordinate-wise descent. We use simulations to compare our method to others in the literature as well as the aforementioned TRW estimator.

Challenges in Species Tree Estimation Under the Multispecies Coalescent Model

PubMed Central

Xu, Bo; Yang, Ziheng

2016-01-01

The multispecies coalescent (MSC) model has emerged as a powerful framework for inferring species phylogenies while accounting for ancestral polymorphism and gene tree-species tree conflict. A number of methods have been developed in the past few years to estimate the species tree under the MSC. The full likelihood methods (including maximum likelihood and Bayesian inference) average over the unknown gene trees and accommodate their uncertainties properly but involve intensive computation. The approximate or summary coalescent methods are computationally fast and are applicable to genomic datasets with thousands of loci, but do not make an efficient use of information in the multilocus data. Most of them take the two-step approach of reconstructing the gene trees for multiple loci by phylogenetic methods and then treating the estimated gene trees as observed data, without accounting for their uncertainties appropriately. In this article we review the statistical nature of the species tree estimation problem under the MSC, and explore the conceptual issues and challenges of species tree estimation by focusing mainly on simple cases of three or four closely related species. We use mathematical analysis and computer simulation to demonstrate that large differences in statistical performance may exist between the two classes of methods. We illustrate that several counterintuitive behaviors may occur with the summary methods but they are due to inefficient use of information in the data by summary methods and vanish when the data are analyzed using full-likelihood methods. These include (i) unidentifiability of parameters in the model, (ii) inconsistency in the so-called anomaly zone, (iii) singularity on the likelihood surface, and (iv) deterioration of performance upon addition of more data. We discuss the challenges and strategies of species tree inference for distantly related species when the molecular clock is violated, and highlight the need for improving the computational efficiency and model realism of the likelihood methods as well as the statistical efficiency of the summary methods. PMID:27927902

Molecular investigations of Hepatozoon species in dogs and developmental stages of Rhipicephalus sanguineus.

PubMed

Aktas, Munir; Ozübek, Sezayi; Ipek, Duygu Neval Sayın

2013-06-01

The occurrence and distribution of Hepatozoon species in stray dogs, and the developmental stages of Rhipicephalus sanguineus detached from the same dogs in Diyarbakır Province, Turkey is reported. A total of 328 ticks, including 133 adults (55 males and 75 females consist of 63 partially engorged and 15 fully engorged) and 195 nymphs (91 partially engorged and 104 fully engorged) were detached from the dogs. Fully engorged nymphs and females were incubated at 27 °C and relative humidity of 85 % to molt to adult stage and recover eggs. The ticks were pooled according to sex and developmental stage. No Hepatozoon gamonts were found, whereas, by PCR, 15.87 % (10/63) of the dogs were infected with Hepatozoon canis. Of the 68 tick pools tested, 14 (20.58 %) pools were infected with Hepatozoon spp., an overall maximum likelihood estimation of prevalence of 4.9 % (95 % confidence intervals (CI) = 2.85-7.93 %) per 100 ticks. Maximum likelihood estimation of the infection rate varied by tick sex and developmental categories, ranging from 1.75 % (95 % CI = 0.11-8.11 %) in fed males to 6.81 % (95 % CI = 2.07-17.46 %) in unfed females. One amplicon from a fed adult female was 99 % identical to the sequence for Hepatozoon felis. The remaining sequences isolated from both dogs and ticks shared 99-100 % similarity with the corresponding H. canis isolates. This is the first detection of H. canis and H. felis in the tick R. sanguineus in Turkey.

Predictive model for risk of cesarean section in pregnant women after induction of labor.

PubMed

Hernández-Martínez, Antonio; Pascual-Pedreño, Ana I; Baño-Garnés, Ana B; Melero-Jiménez, María R; Tenías-Burillo, José M; Molina-Alarcón, Milagros

2016-03-01

To develop a predictive model for risk of cesarean section in pregnant women after induction of labor. A retrospective cohort study was conducted of 861 induced labors during 2009, 2010, and 2011 at Hospital "La Mancha-Centro" in Alcázar de San Juan, Spain. Multivariate analysis was used with binary logistic regression and areas under the ROC curves to determine predictive ability. Two predictive models were created: model A predicts the outcome at the time the woman is admitted to the hospital (before the decision to of the method of induction); and model B predicts the outcome at the time the woman is definitely admitted to the labor room. The predictive factors in the final model were: maternal height, body mass index, nulliparity, Bishop score, gestational age, macrosomia, gender of fetus, and the gynecologist's overall cesarean section rate. The predictive ability of model A was 0.77 [95% confidence interval (CI) 0.73-0.80] and model B was 0.79 (95% CI 0.76-0.83). The predictive ability for pregnant women with previous cesarean section with model A was 0.79 (95% CI 0.64-0.94) and with model B was 0.80 (95% CI 0.64-0.96). For a probability of estimated cesarean section ≥80%, the models A and B presented a positive likelihood ratio (+LR) for cesarean section of 22 and 20, respectively. Also, for a likelihood of estimated cesarean section ≤10%, the models A and B presented a +LR for vaginal delivery of 13 and 6, respectively. These predictive models have a good discriminative ability, both overall and for all subgroups studied. This tool can be useful in clinical practice, especially for pregnant women with previous cesarean section and diabetes.

Body weight concerns: Cross-national study and identification of factors related to eating disorders.

PubMed

Silva, Wanderson Roberto da; Santana, Moema de Souza; Maroco, João; Maloa, Benvindo Felismino Samuel; Campos, Juliana Alvares Duarte Bonini

2017-01-01

Body weight concerns are common among individuals with eating disorders, and this construct can be assessed using psychometric instruments. The Weight Concerns Scale (WCS) is commonly used to assess body weight concerns. To evaluate the psychometric properties of the WCS with Brazilian, Portuguese, and Mozambican female college students; to estimate body weight concerns; and to identify factors related to eating disorders. Confirmatory factor analysis was performed. Factorial, convergent, concurrent, and divergent validity, as well as reliability, were assessed. Cross-national invariance was tested by means of multigroup analysis. Structural models were tested using the WCS as the dependent variable, while demographic and academic variables and body mass index were used as independent variables. Logistic models were tested to estimate the likelihood of eating disorders being developed in specific groups. Participants were 2,068 female students. The psychometric properties of the WCS were adequate for the Portuguese sample; however, for the Brazilian and Mozambican samples, it was necessary to correlate the errors of two items to improve model fit. The WCS did not show cross-national invariance. The variables "thoughts about dropping out of college," "medication use because of studies," "medication and supplements use for body change," "body mass index," "socioeconomic status," "age," and "performance in course" were significant predictors of body weight concerns. Overall, 24.4% (95% confidence interval = 22.9-26.7) of the students were likely to develop eating disorders. Students under 21 years old, who use medication and supplements for body change, and who were classified as overweight/obese have increased likelihood of developing eating disorders. The WCS showed good psychometric properties with Brazilian, Portuguese, and Mozambican students; however, it did not show cross-national invariance. We identified important aspects for investigating body weight concerns and factors related to eating disorders.

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.

Long working hours and use of psychotropic medicine: a follow-up study with register linkage.

PubMed

Hannerz, Harald; Albertsen, Karen

2016-03-01

This study aimed to investigate the possibility of a prospective association between long working hours and use of psychotropic medicine. Survey data drawn from random samples of the general working population of Denmark in the time period 1995-2010 were linked to national registers covering all inhabitants. The participants were followed for first occurrence of redeemed prescriptions for psychotropic medicine. The primary analysis included 25,959 observations (19,259 persons) and yielded a total of 2914 new cases of psychotropic drug use in 99,018 person-years at risk. Poisson regression was used to model incidence rates of redeemed prescriptions for psychotropic medicine as a function of working hours (32-40, 41-48, >48 hours/week). The analysis was controlled for gender, age, sample, shift work, and socioeconomic status. A likelihood ratio test was used to test the null hypothesis, which stated that the incidence rates were independent of weekly working hours. The likelihood ratio test did not reject the null hypothesis (P=0.085). The rate ratio (RR) was 1.04 [95% confidence interval (95% CI) 0.94-1.15] for the contrast 41-48 versus 32-40 work hours/week and 1.15 (95% CI 1.02-1.30) for >48 versus 32-40 hours/week. None of the rate ratios that were estimated in the present study were statistically significant after adjustment for multiple testing. However, stratified analyses, in which 30 RR were estimated, generated the hypothesis that overtime work (>48 hours/week) might be associated with an increased risk among night or shift workers (RR=1.51, 95% CI 1.15-1.98). The present study did not find a statistically significant association between long working hours and incidence of psychotropic drug usage among Danish employees.

Uveitis Reactivation in Children Treated with Tumor Necrosis Factor-α Inhibitors

PubMed Central

Lerman, Melissa A.; Lewen, Michael D.; Kempen, John H.; Mills, Monte D.

2016-01-01

PURPOSE To evaluate reactivation of pediatric uveitis during/following treatment with TNF-alpha inhibition (anti-TNFα). DESIGN Retrospective cohort study. METHODS We assessed the incidence of uveitis reactivation in children ≤18 years who had achieved uveitis quiescence under anti-TNFα. Survival analysis was used to calculate reactivation rates while still on (primary outcome), and following discontinuation of (secondary outcome), anti-TNFα. Potential predictive factors were assessed. RESULTS Among 50 children observed to develop quiescence of uveitis under anti-TNFα, 39 met criteria to be “at risk” of the primary (19 for the secondary) outcome. 60% were female, ~half had Juvenile Idiopathic Arthritis, and most were treated with infliximab. Overall, the estimated proportion relapsing within 12 months was 27.8% (95% confidence interval [CI]: 15.9-45.8%); the estimated probability of reactivation was higher following (63.8% [95% CI: 38.9-87.7%]), than before (21.6% [95% CI: 10.8-40.2%]), anti-TNFα discontinuation. Amongst those who discontinued anti-TNFα, the likelihood of reactivation was higher for those treated with adalimumab vs. infliximab (Hazard Ratio [HR] 13.4, p=0.01, 95% CI: 2.2-82.5) and those with older age at uveitis-onset (HR 1.3, p=0.09, 95% CI: 1.0-1.7). The duration of suppression, on medication, did not significantly affect the likelihood of reactivation when quiescence was maintained for ≥1.5 years. CONCLUSIONS Approximately 75% of children remaining on anti-TNFα following achievement of uveitis quiescence remain quiescent at one year. However, most reactivate following anti-TNFα discontinuation. These results suggest that infliximab more often is followed by remission, off medication, than adalimumab. The data do not suggest that maintenance of suppression, for more than 1.5 years decreases the reactivation risk. PMID:25892124

Sleep Disruption Medical Intervention Forecasting (SDMIF) Module for the Integrated Medical Model

NASA Technical Reports Server (NTRS)

Lewandowski, Beth; Brooker, John; Mallis, Melissa; Hursh, Steve; Caldwell, Lynn; Myers, Jerry

2011-01-01

The NASA Integrated Medical Model (IMM) assesses the risk, including likelihood and impact of occurrence, of all credible in-flight medical conditions. Fatigue due to sleep disruption is a condition that could lead to operational errors, potentially resulting in loss of mission or crew. Pharmacological consumables are mitigation strategies used to manage the risks associated with sleep deficits. The likelihood of medical intervention due to sleep disruption was estimated with a well validated sleep model and a Monte Carlo computer simulation in an effort to optimize the quantity of consumables. METHODS: The key components of the model are the mission parameter program, the calculation of sleep intensity and the diagnosis and decision module. The mission parameter program was used to create simulated daily sleep/wake schedules for an ISS increment. The hypothetical schedules included critical events such as dockings and extravehicular activities and included actual sleep time and sleep quality. The schedules were used as inputs to the Sleep, Activity, Fatigue and Task Effectiveness (SAFTE) Model (IBR Inc., Baltimore MD), which calculated sleep intensity. Sleep data from an ISS study was used to relate calculated sleep intensity to the probability of sleep medication use, using a generalized linear model for binomial regression. A human yes/no decision process using a binomial random number was also factored into sleep medication use probability. RESULTS: These probability calculations were repeated 5000 times resulting in an estimate of the most likely amount of sleep aids used during an ISS mission and a 95% confidence interval. CONCLUSIONS: These results were transferred to the parent IMM for further weighting and integration with other medical conditions, to help inform operational decisions. This model is a potential planning tool for ensuring adequate sleep during sleep disrupted periods of a mission.

A pseudo-penalized quasi-likelihood approach to the spatial misalignment problem with non-normal data.

PubMed

Lopiano, Kenneth K; Young, Linda J; Gotway, Carol A

2014-09-01

Spatially referenced datasets arising from multiple sources are routinely combined to assess relationships among various outcomes and covariates. The geographical units associated with the data, such as the geographical coordinates or areal-level administrative units, are often spatially misaligned, that is, observed at different locations or aggregated over different geographical units. As a result, the covariate is often predicted at the locations where the response is observed. The method used to align disparate datasets must be accounted for when subsequently modeling the aligned data. Here we consider the case where kriging is used to align datasets in point-to-point and point-to-areal misalignment problems when the response variable is non-normally distributed. If the relationship is modeled using generalized linear models, the additional uncertainty induced from using the kriging mean as a covariate introduces a Berkson error structure. In this article, we develop a pseudo-penalized quasi-likelihood algorithm to account for the additional uncertainty when estimating regression parameters and associated measures of uncertainty. The method is applied to a point-to-point example assessing the relationship between low-birth weights and PM2.5 levels after the onset of the largest wildfire in Florida history, the Bugaboo scrub fire. A point-to-areal misalignment problem is presented where the relationship between asthma events in Florida's counties and PM2.5 levels after the onset of the fire is assessed. Finally, the method is evaluated using a simulation study. Our results indicate the method performs well in terms of coverage for 95% confidence intervals and naive methods that ignore the additional uncertainty tend to underestimate the variability associated with parameter estimates. The underestimation is most profound in Poisson regression models. © 2014, The International Biometric Society.

Decoding fMRI events in sensorimotor motor network using sparse paradigm free mapping and activation likelihood estimates.

PubMed

Tan, Francisca M; Caballero-Gaudes, César; Mullinger, Karen J; Cho, Siu-Yeung; Zhang, Yaping; Dryden, Ian L; Francis, Susan T; Gowland, Penny A

2017-11-01

Most functional MRI (fMRI) studies map task-driven brain activity using a block or event-related paradigm. Sparse paradigm free mapping (SPFM) can detect the onset and spatial distribution of BOLD events in the brain without prior timing information, but relating the detected events to brain function remains a challenge. In this study, we developed a decoding method for SPFM using a coordinate-based meta-analysis method of activation likelihood estimation (ALE). We defined meta-maps of statistically significant ALE values that correspond to types of events and calculated a summation overlap between the normalized meta-maps and SPFM maps. As a proof of concept, this framework was applied to relate SPFM-detected events in the sensorimotor network (SMN) to six motor functions (left/right fingers, left/right toes, swallowing, and eye blinks). We validated the framework using simultaneous electromyography (EMG)-fMRI experiments and motor tasks with short and long duration, and random interstimulus interval. The decoding scores were considerably lower for eye movements relative to other movement types tested. The average successful rate for short and long motor events were 77 ± 13% and 74 ± 16%, respectively, excluding eye movements. We found good agreement between the decoding results and EMG for most events and subjects, with a range in sensitivity between 55% and 100%, excluding eye movements. The proposed method was then used to classify the movement types of spontaneous single-trial events in the SMN during resting state, which produced an average successful rate of 22 ± 12%. Finally, this article discusses methodological implications and improvements to increase the decoding performance. Hum Brain Mapp 38:5778-5794, 2017. © 2017 Wiley Periodicals, Inc. © 2017 Wiley Periodicals, Inc.

Decoding fMRI events in Sensorimotor Motor Network using Sparse Paradigm Free Mapping and Activation Likelihood Estimates

PubMed Central

Tan, Francisca M.; Caballero-Gaudes, César; Mullinger, Karen J.; Cho, Siu-Yeung; Zhang, Yaping; Dryden, Ian L.; Francis, Susan T.; Gowland, Penny A.

2017-01-01

Most fMRI studies map task-driven brain activity using a block or event-related paradigm. Sparse Paradigm Free Mapping (SPFM) can detect the onset and spatial distribution of BOLD events in the brain without prior timing information; but relating the detected events to brain function remains a challenge. In this study, we developed a decoding method for SPFM using a coordinate-based meta-analysis method of Activation Likelihood Estimation (ALE). We defined meta-maps of statistically significant ALE values that correspond to types of events and calculated a summation overlap between the normalized meta-maps and SPFM maps. As a proof of concept, this framework was applied to relate SPFM-detected events in the Sensorimotor Network (SMN) to six motor function (left/right fingers, left/right toes, swallowing and eye blinks). We validated the framework using simultaneous Electromyography-fMRI experiments and motor tasks with short and long duration, and random inter-stimulus interval. The decoding scores were considerably lower for eye movements relative to other movement types tested. The average successful rate for short and long motor events was 77 ± 13% and 74 ± 16% respectively, excluding eye movements. We found good agreement between the decoding results and EMG for most events and subjects, with a range in sensitivity between 55 and 100%, excluding eye movements. The proposed method was then used to classify the movement types of spontaneous single-trial events in the SMN during resting state, which produced an average successful rate of 22 ± 12%. Finally, this paper discusses methodological implications and improvements to increase the decoding performance. PMID:28815863

An optimization based sampling approach for multiple metrics uncertainty analysis using generalized likelihood uncertainty estimation

NASA Astrophysics Data System (ADS)

Zhou, Rurui; Li, Yu; Lu, Di; Liu, Haixing; Zhou, Huicheng

2016-09-01

This paper investigates the use of an epsilon-dominance non-dominated sorted genetic algorithm II (ɛ-NSGAII) as a sampling approach with an aim to improving sampling efficiency for multiple metrics uncertainty analysis using Generalized Likelihood Uncertainty Estimation (GLUE). The effectiveness of ɛ-NSGAII based sampling is demonstrated compared with Latin hypercube sampling (LHS) through analyzing sampling efficiency, multiple metrics performance, parameter uncertainty and flood forecasting uncertainty with a case study of flood forecasting uncertainty evaluation based on Xinanjiang model (XAJ) for Qing River reservoir, China. Results obtained demonstrate the following advantages of the ɛ-NSGAII based sampling approach in comparison to LHS: (1) The former performs more effective and efficient than LHS, for example the simulation time required to generate 1000 behavioral parameter sets is shorter by 9 times; (2) The Pareto tradeoffs between metrics are demonstrated clearly with the solutions from ɛ-NSGAII based sampling, also their Pareto optimal values are better than those of LHS, which means better forecasting accuracy of ɛ-NSGAII parameter sets; (3) The parameter posterior distributions from ɛ-NSGAII based sampling are concentrated in the appropriate ranges rather than uniform, which accords with their physical significance, also parameter uncertainties are reduced significantly; (4) The forecasted floods are close to the observations as evaluated by three measures: the normalized total flow outside the uncertainty intervals (FOUI), average relative band-width (RB) and average deviation amplitude (D). The flood forecasting uncertainty is also reduced a lot with ɛ-NSGAII based sampling. This study provides a new sampling approach to improve multiple metrics uncertainty analysis under the framework of GLUE, and could be used to reveal the underlying mechanisms of parameter sets under multiple conflicting metrics in the uncertainty analysis process.

COSMIC MICROWAVE BACKGROUND LIKELIHOOD APPROXIMATION FOR BANDED PROBABILITY DISTRIBUTIONS

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

Gjerløw, E.; Mikkelsen, K.; Eriksen, H. K.

We investigate sets of random variables that can be arranged sequentially such that a given variable only depends conditionally on its immediate predecessor. For such sets, we show that the full joint probability distribution may be expressed exclusively in terms of uni- and bivariate marginals. Under the assumption that the cosmic microwave background (CMB) power spectrum likelihood only exhibits correlations within a banded multipole range, Δl{sub C}, we apply this expression to two outstanding problems in CMB likelihood analysis. First, we derive a statistically well-defined hybrid likelihood estimator, merging two independent (e.g., low- and high-l) likelihoods into a single expressionmore » that properly accounts for correlations between the two. Applying this expression to the Wilkinson Microwave Anisotropy Probe (WMAP) likelihood, we verify that the effect of correlations on cosmological parameters in the transition region is negligible in terms of cosmological parameters for WMAP; the largest relative shift seen for any parameter is 0.06σ. However, because this may not hold for other experimental setups (e.g., for different instrumental noise properties or analysis masks), but must rather be verified on a case-by-case basis, we recommend our new hybridization scheme for future experiments for statistical self-consistency reasons. Second, we use the same expression to improve the convergence rate of the Blackwell-Rao likelihood estimator, reducing the required number of Monte Carlo samples by several orders of magnitude, and thereby extend it to high-l applications.« less

Development of the Average Likelihood Function for Code Division Multiple Access (CDMA) Using BPSK and QPSK Symbols

DTIC Science & Technology

2015-01-01

This research has the purpose to establish a foundation for new classification and estimation of CDMA signals. Keywords: DS / CDMA signals, BPSK, QPSK...DEVELOPMENT OF THE AVERAGE LIKELIHOOD FUNCTION FOR CODE DIVISION MULTIPLE ACCESS ( CDMA ) USING BPSK AND QPSK SYMBOLS JANUARY 2015...To) OCT 2013 – OCT 2014 4. TITLE AND SUBTITLE DEVELOPMENT OF THE AVERAGE LIKELIHOOD FUNCTION FOR CODE DIVISION MULTIPLE ACCESS ( CDMA ) USING BPSK

Doubly Robust and Efficient Estimation of Marginal Structural Models for the Hazard Function

PubMed Central

Zheng, Wenjing; Petersen, Maya; van der Laan, Mark

2016-01-01

In social and health sciences, many research questions involve understanding the causal effect of a longitudinal treatment on mortality (or time-to-event outcomes in general). Often, treatment status may change in response to past covariates that are risk factors for mortality, and in turn, treatment status may also affect such subsequent covariates. In these situations, Marginal Structural Models (MSMs), introduced by Robins (1997), are well-established and widely used tools to account for time-varying confounding. In particular, a MSM can be used to specify the intervention-specific counterfactual hazard function, i.e. the hazard for the outcome of a subject in an ideal experiment where he/she was assigned to follow a given intervention on their treatment variables. The parameters of this hazard MSM are traditionally estimated using the Inverse Probability Weighted estimation (IPTW, van der Laan and Petersen (2007), Robins et al. (2000b), Robins (1999), Robins et al. (2008)). This estimator is easy to implement and admits Wald-type confidence intervals. However, its consistency hinges on the correct specification of the treatment allocation probabilities, and the estimates are generally sensitive to large treatment weights (especially in the presence of strong confounding), which are difficult to stabilize for dynamic treatment regimes. In this paper, we present a pooled targeted maximum likelihood estimator (TMLE, van der Laan and Rubin (2006)) for MSM for the hazard function under longitudinal dynamic treatment regimes. The proposed estimator is semiparametric efficient and doubly robust, hence offers bias reduction and efficiency gain over the incumbent IPTW estimator. Moreover, the substitution principle rooted in the TMLE potentially mitigates the sensitivity to large treatment weights in IPTW. We compare the performance of the proposed estimator with the IPTW and a non-targeted substitution estimator in a simulation study. PMID:27227723

Estimating temporary emigration and breeding proportions using capture-recapture data with Pollock's robust design

USGS Publications Warehouse

Kendall, W.L.; Nichols, J.D.; Hines, J.E.

1997-01-01

Statistical inference for capture-recapture studies of open animal populations typically relies on the assumption that all emigration from the studied population is permanent. However, there are many instances in which this assumption is unlikely to be met. We define two general models for the process of temporary emigration, completely random and Markovian. We then consider effects of these two types of temporary emigration on Jolly-Seber (Seber 1982) estimators and on estimators arising from the full-likelihood approach of Kendall et al. (1995) to robust design data. Capture-recapture data arising from Pollock's (1982) robust design provide the basis for obtaining unbiased estimates of demographic parameters in the presence of temporary emigration and for estimating the probability of temporary emigration. We present a likelihood-based approach to dealing with temporary emigration that permits estimation under different models of temporary emigration and yields tests for completely random and Markovian emigration. In addition, we use the relationship between capture probability estimates based on closed and open models under completely random temporary emigration to derive three ad hoc estimators for the probability of temporary emigration, two of which should be especially useful in situations where capture probabilities are heterogeneous among individual animals. Ad hoc and full-likelihood estimators are illustrated for small mammal capture-recapture data sets. We believe that these models and estimators will be useful for testing hypotheses about the process of temporary emigration, for estimating demographic parameters in the presence of temporary emigration, and for estimating probabilities of temporary emigration. These latter estimates are frequently of ecological interest as indicators of animal movement and, in some sampling situations, as direct estimates of breeding probabilities and proportions.

Modified Maxium Likelihood Estimation Method for Completely Separated and Quasi-Completely Separated Data for a Dose-Response Model

DTIC Science & Technology

2015-08-01

McCullagh, P.; Nelder, J.A. Generalized Linear Model , 2nd ed.; Chapman and Hall: London, 1989. 7. Johnston, J. Econometric Methods, 3rd ed.; McGraw...FOR A DOSE-RESPONSE MODEL ECBC-TN-068 Kyong H. Park Steven J. Lagan RESEARCH AND TECHNOLOGY DIRECTORATE August 2015 Approved for public release...Likelihood Estimation Method for Completely Separated and Quasi-Completely Separated Data for a Dose-Response Model 5a. CONTRACT NUMBER 5b. GRANT

Zero-inflated Poisson model based likelihood ratio test for drug safety signal detection.

PubMed

Huang, Lan; Zheng, Dan; Zalkikar, Jyoti; Tiwari, Ram

2017-02-01

In recent decades, numerous methods have been developed for data mining of large drug safety databases, such as Food and Drug Administration's (FDA's) Adverse Event Reporting System, where data matrices are formed by drugs such as columns and adverse events as rows. Often, a large number of cells in these data matrices have zero cell counts and some of them are "true zeros" indicating that the drug-adverse event pairs cannot occur, and these zero counts are distinguished from the other zero counts that are modeled zero counts and simply indicate that the drug-adverse event pairs have not occurred yet or have not been reported yet. In this paper, a zero-inflated Poisson model based likelihood ratio test method is proposed to identify drug-adverse event pairs that have disproportionately high reporting rates, which are also called signals. The maximum likelihood estimates of the model parameters of zero-inflated Poisson model based likelihood ratio test are obtained using the expectation and maximization algorithm. The zero-inflated Poisson model based likelihood ratio test is also modified to handle the stratified analyses for binary and categorical covariates (e.g. gender and age) in the data. The proposed zero-inflated Poisson model based likelihood ratio test method is shown to asymptotically control the type I error and false discovery rate, and its finite sample performance for signal detection is evaluated through a simulation study. The simulation results show that the zero-inflated Poisson model based likelihood ratio test method performs similar to Poisson model based likelihood ratio test method when the estimated percentage of true zeros in the database is small. Both the zero-inflated Poisson model based likelihood ratio test and likelihood ratio test methods are applied to six selected drugs, from the 2006 to 2011 Adverse Event Reporting System database, with varying percentages of observed zero-count cells.

Epidemiologic programs for computers and calculators. A microcomputer program for multiple logistic regression by unconditional and conditional maximum likelihood methods.

PubMed

Campos-Filho, N; Franco, E L

1989-02-01

A frequent procedure in matched case-control studies is to report results from the multivariate unmatched analyses if they do not differ substantially from the ones obtained after conditioning on the matching variables. Although conceptually simple, this rule requires that an extensive series of logistic regression models be evaluated by both the conditional and unconditional maximum likelihood methods. Most computer programs for logistic regression employ only one maximum likelihood method, which requires that the analyses be performed in separate steps. This paper describes a Pascal microcomputer (IBM PC) program that performs multiple logistic regression by both maximum likelihood estimation methods, which obviates the need for switching between programs to obtain relative risk estimates from both matched and unmatched analyses. The program calculates most standard statistics and allows factoring of categorical or continuous variables by two distinct methods of contrast. A built-in, descriptive statistics option allows the user to inspect the distribution of cases and controls across categories of any given variable.

Interval stability for complex systems

NASA Astrophysics Data System (ADS)

Klinshov, Vladimir V.; Kirillov, Sergey; Kurths, Jürgen; Nekorkin, Vladimir I.

2018-04-01

Stability of dynamical systems against strong perturbations is an important problem of nonlinear dynamics relevant to many applications in various areas. Here, we develop a novel concept of interval stability, referring to the behavior of the perturbed system during a finite time interval. Based on this concept, we suggest new measures of stability, namely interval basin stability (IBS) and interval stability threshold (IST). IBS characterizes the likelihood that the perturbed system returns to the stable regime (attractor) in a given time. IST provides the minimal magnitude of the perturbation capable to disrupt the stable regime for a given interval of time. The suggested measures provide important information about the system susceptibility to external perturbations which may be useful for practical applications. Moreover, from a theoretical viewpoint the interval stability measures are shown to bridge the gap between linear and asymptotic stability. We also suggest numerical algorithms for quantification of the interval stability characteristics and demonstrate their potential for several dynamical systems of various nature, such as power grids and neural networks.

Background stratified Poisson regression analysis of cohort data.

PubMed

Richardson, David B; Langholz, Bryan

2012-03-01

Background stratified Poisson regression is an approach that has been used in the analysis of data derived from a variety of epidemiologically important studies of radiation-exposed populations, including uranium miners, nuclear industry workers, and atomic bomb survivors. We describe a novel approach to fit Poisson regression models that adjust for a set of covariates through background stratification while directly estimating the radiation-disease association of primary interest. The approach makes use of an expression for the Poisson likelihood that treats the coefficients for stratum-specific indicator variables as 'nuisance' variables and avoids the need to explicitly estimate the coefficients for these stratum-specific parameters. Log-linear models, as well as other general relative rate models, are accommodated. This approach is illustrated using data from the Life Span Study of Japanese atomic bomb survivors and data from a study of underground uranium miners. The point estimate and confidence interval obtained from this 'conditional' regression approach are identical to the values obtained using unconditional Poisson regression with model terms for each background stratum. Moreover, it is shown that the proposed approach allows estimation of background stratified Poisson regression models of non-standard form, such as models that parameterize latency effects, as well as regression models in which the number of strata is large, thereby overcoming the limitations of previously available statistical software for fitting background stratified Poisson regression models.

A Game Theoretical Approach to Hacktivism: Is Attack Likelihood a Product of Risks and Payoffs?

PubMed

Bodford, Jessica E; Kwan, Virginia S Y

2018-02-01

The current study examines hacktivism (i.e., hacking to convey a moral, ethical, or social justice message) through a general game theoretic framework-that is, as a product of costs and benefits. Given the inherent risk of carrying out a hacktivist attack (e.g., legal action, imprisonment), it would be rational for the user to weigh these risks against perceived benefits of carrying out the attack. As such, we examined computer science students' estimations of risks, payoffs, and attack likelihood through a game theoretic design. Furthermore, this study aims at constructing a descriptive profile of potential hacktivists, exploring two predicted covariates of attack decision making, namely, peer prevalence of hacking and sex differences. Contrary to expectations, results suggest that participants' estimations of attack likelihood stemmed solely from expected payoffs, rather than subjective risks. Peer prevalence significantly predicted increased payoffs and attack likelihood, suggesting an underlying descriptive norm in social networks. Notably, we observed no sex differences in the decision to attack, nor in the factors predicting attack likelihood. Implications for policymakers and the understanding and prevention of hacktivism are discussed, as are the possible ramifications of widely communicated payoffs over potential risks in hacking communities.

Uncertainty estimation of Intensity-Duration-Frequency relationships: A regional analysis

NASA Astrophysics Data System (ADS)

Mélèse, Victor; Blanchet, Juliette; Molinié, Gilles

2018-03-01

We propose in this article a regional study of uncertainties in IDF curves derived from point-rainfall maxima. We develop two generalized extreme value models based on the simple scaling assumption, first in the frequentist framework and second in the Bayesian framework. Within the frequentist framework, uncertainties are obtained i) from the Gaussian density stemming from the asymptotic normality theorem of the maximum likelihood and ii) with a bootstrap procedure. Within the Bayesian framework, uncertainties are obtained from the posterior densities. We confront these two frameworks on the same database covering a large region of 100, 000 km2 in southern France with contrasted rainfall regime, in order to be able to draw conclusion that are not specific to the data. The two frameworks are applied to 405 hourly stations with data back to the 1980's, accumulated in the range 3 h-120 h. We show that i) the Bayesian framework is more robust than the frequentist one to the starting point of the estimation procedure, ii) the posterior and the bootstrap densities are able to better adjust uncertainty estimation to the data than the Gaussian density, and iii) the bootstrap density give unreasonable confidence intervals, in particular for return levels associated to large return period. Therefore our recommendation goes towards the use of the Bayesian framework to compute uncertainty.

Adaptive pre-specification in randomized trials with and without pair-matching.

PubMed

Balzer, Laura B; van der Laan, Mark J; Petersen, Maya L

2016-11-10

In randomized trials, adjustment for measured covariates during the analysis can reduce variance and increase power. To avoid misleading inference, the analysis plan must be pre-specified. However, it is often unclear a priori which baseline covariates (if any) should be adjusted for in the analysis. Consider, for example, the Sustainable East Africa Research in Community Health (SEARCH) trial for HIV prevention and treatment. There are 16 matched pairs of communities and many potential adjustment variables, including region, HIV prevalence, male circumcision coverage, and measures of community-level viral load. In this paper, we propose a rigorous procedure to data-adaptively select the adjustment set, which maximizes the efficiency of the analysis. Specifically, we use cross-validation to select from a pre-specified library the candidate targeted maximum likelihood estimator (TMLE) that minimizes the estimated variance. For further gains in precision, we also propose a collaborative procedure for estimating the known exposure mechanism. Our small sample simulations demonstrate the promise of the methodology to maximize study power, while maintaining nominal confidence interval coverage. We show how our procedure can be tailored to the scientific question (intervention effect for the study sample vs. for the target population) and study design (pair-matched or not). Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

A Statistical Atrioventricular Node Model Accounting for Pathway Switching During Atrial Fibrillation.

PubMed

Henriksson, Mikael; Corino, Valentina D A; Sornmo, Leif; Sandberg, Frida

2016-09-01

The atrioventricular (AV) node plays a central role in atrial fibrillation (AF), as it influences the conduction of impulses from the atria into the ventricles. In this paper, the statistical dual pathway AV node model, previously introduced by us, is modified so that it accounts for atrial impulse pathway switching even if the preceding impulse did not cause a ventricular activation. The proposed change in model structure implies that the number of model parameters subjected to maximum likelihood estimation is reduced from five to four. The model is evaluated using the data acquired in the RATe control in atrial fibrillation (RATAF) study, involving 24-h ECG recordings from 60 patients with permanent AF. When fitting the models to the RATAF database, similar results were obtained for both the present and the previous model, with a median fit of 86%. The results show that the parameter estimates characterizing refractory period prolongation exhibit considerably lower variation when using the present model, a finding that may be ascribed to fewer model parameters. The new model maintains the capability to model RR intervals, while providing more reliable parameters estimates. The model parameters are expected to convey novel clinical information, and may be useful for predicting the effect of rate control drugs.

Phylogeography of the Western Lyresnake (Trimorphodon biscutatus): testing aridland biogeographical hypotheses across the Nearctic-Neotropical transition.

PubMed

Devitt, Thomas J

2006-12-01

The Western Lyresnake (Trimorphodon biscutatus) is a widespread, polytypic taxon inhabiting arid regions from the warm deserts of the southwestern United States southward along the Pacific versant of Mexico to the tropical deciduous forests of Mesoamerica. This broadly distributed species provides a unique opportunity to evaluate a priori biogeographical hypotheses spanning two major distinct biogeographical realms (the Nearctic and Neotropical) that are usually treated separately in phylogeographical analyses. I investigated the phylogeography of T. biscutatus using maximum likelihood and Bayesian phylogenetic analysis of mitochondrial DNA (mtDNA) from across this species' range. Phylogenetic analyses recovered five well-supported clades whose boundaries are concordant with existing geographical barriers, a pattern consistent with a model of vicariant allopatric divergence. Assuming a vicariance model, divergence times between mitochondrial lineages were estimated using Bayesian relaxed molecular clock methods calibrated using geological information from putative vicariant events. Divergence time point estimates were bounded by broad confidence intervals, and thus these highly conservative estimates should be considered tentative hypotheses at best. Comparison of mtDNA lineages and taxa traditionally recognized as subspecies based on morphology suggest this taxon is comprised of multiple independent lineages at various stages of divergence, ranging from putative secondary contact and hybridization to sympatry of 'subspecies'.

Accounting for Sampling Error in Genetic Eigenvalues Using Random Matrix Theory.

PubMed

Sztepanacz, Jacqueline L; Blows, Mark W

2017-07-01

The distribution of genetic variance in multivariate phenotypes is characterized by the empirical spectral distribution of the eigenvalues of the genetic covariance matrix. Empirical estimates of genetic eigenvalues from random effects linear models are known to be overdispersed by sampling error, where large eigenvalues are biased upward, and small eigenvalues are biased downward. The overdispersion of the leading eigenvalues of sample covariance matrices have been demonstrated to conform to the Tracy-Widom (TW) distribution. Here we show that genetic eigenvalues estimated using restricted maximum likelihood (REML) in a multivariate random effects model with an unconstrained genetic covariance structure will also conform to the TW distribution after empirical scaling and centering. However, where estimation procedures using either REML or MCMC impose boundary constraints, the resulting genetic eigenvalues tend not be TW distributed. We show how using confidence intervals from sampling distributions of genetic eigenvalues without reference to the TW distribution is insufficient protection against mistaking sampling error as genetic variance, particularly when eigenvalues are small. By scaling such sampling distributions to the appropriate TW distribution, the critical value of the TW statistic can be used to determine if the magnitude of a genetic eigenvalue exceeds the sampling error for each eigenvalue in the spectral distribution of a given genetic covariance matrix. Copyright © 2017 by the Genetics Society of America.

Scanning linear estimation: improvements over region of interest (ROI) methods

NASA Astrophysics Data System (ADS)

Kupinski, Meredith K.; Clarkson, Eric W.; Barrett, Harrison H.

2013-03-01

In tomographic medical imaging, a signal activity is typically estimated by summing voxels from a reconstructed image. We introduce an alternative estimation scheme that operates on the raw projection data and offers a substantial improvement, as measured by the ensemble mean-square error (EMSE), when compared to using voxel values from a maximum-likelihood expectation-maximization (MLEM) reconstruction. The scanning-linear (SL) estimator operates on the raw projection data and is derived as a special case of maximum-likelihood estimation with a series of approximations to make the calculation tractable. The approximated likelihood accounts for background randomness, measurement noise and variability in the parameters to be estimated. When signal size and location are known, the SL estimate of signal activity is unbiased, i.e. the average estimate equals the true value. By contrast, unpredictable bias arising from the null functions of the imaging system affect standard algorithms that operate on reconstructed data. The SL method is demonstrated for two different tasks: (1) simultaneously estimating a signal’s size, location and activity; (2) for a fixed signal size and location, estimating activity. Noisy projection data are realistically simulated using measured calibration data from the multi-module multi-resolution small-animal SPECT imaging system. For both tasks, the same set of images is reconstructed using the MLEM algorithm (80 iterations), and the average and maximum values within the region of interest (ROI) are calculated for comparison. This comparison shows dramatic improvements in EMSE for the SL estimates. To show that the bias in ROI estimates affects not only absolute values but also relative differences, such as those used to monitor the response to therapy, the activity estimation task is repeated for three different signal sizes.

Asymptotic formulae for likelihood-based tests of new physics

NASA Astrophysics Data System (ADS)

Cowan, Glen; Cranmer, Kyle; Gross, Eilam; Vitells, Ofer

2011-02-01

We describe likelihood-based statistical tests for use in high energy physics for the discovery of new phenomena and for construction of confidence intervals on model parameters. We focus on the properties of the test procedures that allow one to account for systematic uncertainties. Explicit formulae for the asymptotic distributions of test statistics are derived using results of Wilks and Wald. We motivate and justify the use of a representative data set, called the "Asimov data set", which provides a simple method to obtain the median experimental sensitivity of a search or measurement as well as fluctuations about this expectation.

Comparing Three Estimation Methods for the Three-Parameter Logistic IRT Model

ERIC Educational Resources Information Center

Lamsal, Sunil

2015-01-01

Different estimation procedures have been developed for the unidimensional three-parameter item response theory (IRT) model. These techniques include the marginal maximum likelihood estimation, the fully Bayesian estimation using Markov chain Monte Carlo simulation techniques, and the Metropolis-Hastings Robbin-Monro estimation. With each…

Efficient Robust Regression via Two-Stage Generalized Empirical Likelihood

PubMed Central

Bondell, Howard D.; Stefanski, Leonard A.

2013-01-01

Large- and finite-sample efficiency and resistance to outliers are the key goals of robust statistics. Although often not simultaneously attainable, we develop and study a linear regression estimator that comes close. Efficiency obtains from the estimator’s close connection to generalized empirical likelihood, and its favorable robustness properties are obtained by constraining the associated sum of (weighted) squared residuals. We prove maximum attainable finite-sample replacement breakdown point, and full asymptotic efficiency for normal errors. Simulation evidence shows that compared to existing robust regression estimators, the new estimator has relatively high efficiency for small sample sizes, and comparable outlier resistance. The estimator is further illustrated and compared to existing methods via application to a real data set with purported outliers. PMID:23976805

A real-time digital program for estimating aircraft stability and control parameters from flight test data by using the maximum likelihood method

NASA Technical Reports Server (NTRS)

Grove, R. D.; Mayhew, S. C.

1973-01-01

A computer program (Langley program C1123) has been developed for estimating aircraft stability and control parameters from flight test data. These parameters are estimated by the maximum likelihood estimation procedure implemented on a real-time digital simulation system, which uses the Control Data 6600 computer. This system allows the investigator to interact with the program in order to obtain satisfactory results. Part of this system, the control and display capabilities, is described for this program. This report also describes the computer program by presenting the program variables, subroutines, flow charts, listings, and operational features. Program usage is demonstrated with a test case using pseudo or simulated flight data.

Nonparametric Discrete Survival Function Estimation with Uncertain Endpoints Using an Internal Validation Subsample

PubMed Central

Zee, Jarcy; Xie, Sharon X.

2015-01-01

Summary When a true survival endpoint cannot be assessed for some subjects, an alternative endpoint that measures the true endpoint with error may be collected, which often occurs when obtaining the true endpoint is too invasive or costly. We develop an estimated likelihood function for the situation where we have both uncertain endpoints for all participants and true endpoints for only a subset of participants. We propose a nonparametric maximum estimated likelihood estimator of the discrete survival function of time to the true endpoint. We show that the proposed estimator is consistent and asymptotically normal. We demonstrate through extensive simulations that the proposed estimator has little bias compared to the naïve Kaplan-Meier survival function estimator, which uses only uncertain endpoints, and more efficient with moderate missingness compared to the complete-case Kaplan-Meier survival function estimator, which uses only available true endpoints. Finally, we apply the proposed method to a dataset for estimating the risk of developing Alzheimer's disease from the Alzheimer's Disease Neuroimaging Initiative. PMID:25916510

Maximum likelihood phase-retrieval algorithm: applications.

PubMed

Nahrstedt, D A; Southwell, W H

1984-12-01

The maximum likelihood estimator approach is shown to be effective in determining the wave front aberration in systems involving laser and flow field diagnostics and optical testing. The robustness of the algorithm enables convergence even in cases of severe wave front error and real, nonsymmetrical, obscured amplitude distributions.

Cramer-Rao Bound, MUSIC, and Maximum Likelihood. Effects of Temporal Phase Difference

DTIC Science & Technology

1990-11-01

Technical Report 1373 November 1990 Cramer-Rao Bound, MUSIC , And Maximum Likelihood Effects of Temporal Phase o Difference C. V. TranI OTIC Approved... MUSIC , and Maximum Likelihood (ML) asymptotic variances corresponding to the two-source direction-of-arrival estimation where sources were modeled as...1pI = 1.00, SNR = 20 dB ..................................... 27 2. MUSIC for two equipowered signals impinging on a 5-element ULA (a) IpI = 0.50, SNR

Bayesian experimental design for models with intractable likelihoods.

PubMed

Drovandi, Christopher C; Pettitt, Anthony N

2013-12-01

In this paper we present a methodology for designing experiments for efficiently estimating the parameters of models with computationally intractable likelihoods. The approach combines a commonly used methodology for robust experimental design, based on Markov chain Monte Carlo sampling, with approximate Bayesian computation (ABC) to ensure that no likelihood evaluations are required. The utility function considered for precise parameter estimation is based upon the precision of the ABC posterior distribution, which we form efficiently via the ABC rejection algorithm based on pre-computed model simulations. Our focus is on stochastic models and, in particular, we investigate the methodology for Markov process models of epidemics and macroparasite population evolution. The macroparasite example involves a multivariate process and we assess the loss of information from not observing all variables. © 2013, The International Biometric Society.

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.

Prospective study of vaginal bacterial flora and other risk factors for vulvovaginal candidiasis.

PubMed

McClelland, R Scott; Richardson, Barbra A; Hassan, Wisal M; Graham, Susan M; Kiarie, James; Baeten, Jared M; Mandaliya, Kishorchandra; Jaoko, Walter; Ndinya-Achola, Jeckoniah O; Holmes, King K

2009-06-15

It has been suggested that vaginal colonization with lactobacilli may reduce the risk of vulvovaginal candidiasis (VVC), but supporting data are limited. Our objective was to determine the relationship between vaginal bacterial flora and VVC. We conducted a prospective cohort analysis that involved 151 Kenyan sex workers. At monthly follow-up visits, VVC was defined as the presence of yeast buds, pseudohyphae, or both on a wet preparation (including potassium hydroxide preparation) of vaginal secretions. Generalized estimating equations were used to identify correlates of VVC. Participants returned for a median of 12 visits (interquartile range, 11-12 visits). VVC was identified at 162 visits, including 26 involving symptomatic VVC. Bacterial vaginosis was associated with fewer episodes of VVC (adjusted odds ratio [aOR], 0.29 [95% confidence interval {CI}, 0.16-0.50]). After excluding women with concurrent bacterial vaginosis, another possible cause of vaginal symptoms, the likelihood of symptomatic VVC was higher among those who had had yeast identified on wet preparation of vaginal secretions during the past 60 days (aOR, 4.06 [95% CI, 1.12-14.74]) and those with concurrent vaginal Lactobacillus colonization (aOR, 3.75 [95% CI, 1.30-10.83]). Contrary to the commonly posited hypothesis that vaginal Lactobacillus colonization has a protective effect, we found that such colonization was associated with a nearly 4-fold increase in the likelihood of symptomatic VVC.

A Prospective Study of Vaginal Bacterial Flora and Other Risk Factors for Vulvovaginal Candidiasis

PubMed Central

McClelland, R. Scott; Richardson, Barbra A.; Hassan, Wisal M.; Graham, Susan M.; Kiarie, James; Baeten, Jared M.; Mandaliya, Kishorchandra; Jaoko, Walter; Ndinya-Achola, Jeckoniah O.; Holmes, King K.

2009-01-01

Background: It has been suggested that vaginal lactobacilli may reduce the risk of vulvovaginal candidiasis (VVC), but supporting data are limited. Our objective was to determine the relationship between vaginal bacterial flora and VVC. Methods: We conducted a prospective cohort analysis among 151 Kenyan sex workers. At monthly follow-up, VVC was defined as the presence of yeast buds, pseudohyphae, or both on vaginal wet preparation or KOH preparation. Generalized estimating equations were used to identify correlates of VVC. Results: Participants returned for a median of 12 (interquartile range 11-12) visits. Vulvovaginal candidiasis was present at 162 visits, including 26 with symptomatic VVC. Bacterial vaginosis (BV) was associated with fewer episodes of VVC (adjusted odds ratio [aOR] 0.29, 95% confidence interval [CI] 0.16-0.50). After excluding women with concurrent BV, another possible cause of vaginal symptoms, the likelihood of symptomatic VVC was higher in those with yeast on vaginal wet preparation in the past 60 days (aOR 4.06, 95% CI 1.12-14.74) and those with concurrent vaginal Lactobacillus colonization (aOR 3.75, 95% CI 1.30-10.83). Conclusions: Contrary to a commonly posed hypothesis of a protective effect, we found that vaginal Lactobacillus colonization was associated with a >4-fold increase in the likelihood of symptomatic VVC. PMID:19456235

Use of the gamma distribution to represent monthly rainfall in Africa for drought monitoring applications

USGS Publications Warehouse

Husak, Gregory J.; Michaelsen, Joel C.; Funk, Christopher C.

2007-01-01

Evaluating a range of scenarios that accurately reflect precipitation variability is critical for water resource applications. Inputs to these applications can be provided using location- and interval-specific probability distributions. These distributions make it possible to estimate the likelihood of rainfall being within a specified range. In this paper, we demonstrate the feasibility of fitting cell-by-cell probability distributions to grids of monthly interpolated, continent-wide data. Future work will then detail applications of these grids to improved satellite-remote sensing of drought and interpretations of probabilistic climate outlook forum forecasts. The gamma distribution is well suited to these applications because it is fairly familiar to African scientists, and capable of representing a variety of distribution shapes. This study tests the goodness-of-fit using the Kolmogorov–Smirnov (KS) test, and compares these results against another distribution commonly used in rainfall events, the Weibull. The gamma distribution is suitable for roughly 98% of the locations over all months. The techniques and results presented in this study provide a foundation for use of the gamma distribution to generate drivers for various rain-related models. These models are used as decision support tools for the management of water and agricultural resources as well as food reserves by providing decision makers with ways to evaluate the likelihood of various rainfall accumulations and assess different scenarios in Africa. 

Racial/ethnic disparities in self-reported short sleep duration among US-born and foreign-born adults.

PubMed

Cunningham, Timothy J; Wheaton, Anne G; Ford, Earl S; Croft, Janet B

2016-12-01

Racial/ethnic health disparities are infrequently considered by nativity status in the United States, although the immigrant population has practically doubled since 1990. We investigated the modifying role of nativity status (US- vs. foreign-born) on racial/ethnic disparities in short sleep duration (<7 h), which has serious health consequences. Cross-sectional data from 23,505 US-born and 4,326 foreign-born adults aged ≥ 18 years from the 2012 National Health Interview Survey and multivariable log-linear regression were used to estimate prevalence ratios (PR) for reporting short sleep duration and their corresponding 95% confidence intervals (CI). After controlling for sociodemographic covariates, short sleep was more prevalent among blacks (PR 1.29, 95% CI: 1.21-1.37), Hispanics (PR 1.18, 95% CI: 1.08, 1.29), and Asians (PR 1.37, 95% CI: 1.16-1.61) than whites among US-born adults. Short sleep was more prevalent among blacks (PR 1.71, 95% CI: 1.38, 2.13) and Asians (PR 1.23, 95% CI: 1.02, 1.47) than whites among the foreign-born. Among both US- and foreign-born adults, blacks and Asians had a higher likelihood of short sleep compared to whites. US-born Hispanics, but not foreign-born Hispanics, had a higher likelihood than their white counterparts. Future research should aim to uncover mechanisms underlying these disparities.

Fault detection and isolation in GPS receiver autonomous integrity monitoring based on chaos particle swarm optimization-particle filter algorithm

NASA Astrophysics Data System (ADS)

Wang, Ershen; Jia, Chaoying; Tong, Gang; Qu, Pingping; Lan, Xiaoyu; Pang, Tao

2018-03-01

The receiver autonomous integrity monitoring (RAIM) is one of the most important parts in an avionic navigation system. Two problems need to be addressed to improve this system, namely, the degeneracy phenomenon and lack of samples for the standard particle filter (PF). However, the number of samples cannot adequately express the real distribution of the probability density function (i.e., sample impoverishment). This study presents a GPS receiver autonomous integrity monitoring (RAIM) method based on a chaos particle swarm optimization particle filter (CPSO-PF) algorithm with a log likelihood ratio. The chaos sequence generates a set of chaotic variables, which are mapped to the interval of optimization variables to improve particle quality. This chaos perturbation overcomes the potential for the search to become trapped in a local optimum in the particle swarm optimization (PSO) algorithm. Test statistics are configured based on a likelihood ratio, and satellite fault detection is then conducted by checking the consistency between the state estimate of the main PF and those of the auxiliary PFs. Based on GPS data, the experimental results demonstrate that the proposed algorithm can effectively detect and isolate satellite faults under conditions of non-Gaussian measurement noise. Moreover, the performance of the proposed novel method is better than that of RAIM based on the PF or PSO-PF algorithm.

Robust Multi-Frame Adaptive Optics Image Restoration Algorithm Using Maximum Likelihood Estimation with Poisson Statistics.

PubMed

Li, Dongming; Sun, Changming; Yang, Jinhua; Liu, Huan; Peng, Jiaqi; Zhang, Lijuan

2017-04-06

An adaptive optics (AO) system provides real-time compensation for atmospheric turbulence. However, an AO image is usually of poor contrast because of the nature of the imaging process, meaning that the image contains information coming from both out-of-focus and in-focus planes of the object, which also brings about a loss in quality. In this paper, we present a robust multi-frame adaptive optics image restoration algorithm via maximum likelihood estimation. Our proposed algorithm uses a maximum likelihood method with image regularization as the basic principle, and constructs the joint log likelihood function for multi-frame AO images based on a Poisson distribution model. To begin with, a frame selection method based on image variance is applied to the observed multi-frame AO images to select images with better quality to improve the convergence of a blind deconvolution algorithm. Then, by combining the imaging conditions and the AO system properties, a point spread function estimation model is built. Finally, we develop our iterative solutions for AO image restoration addressing the joint deconvolution issue. We conduct a number of experiments to evaluate the performances of our proposed algorithm. Experimental results show that our algorithm produces accurate AO image restoration results and outperforms the current state-of-the-art blind deconvolution methods.

Robust Multi-Frame Adaptive Optics Image Restoration Algorithm Using Maximum Likelihood Estimation with Poisson Statistics

PubMed Central

Li, Dongming; Sun, Changming; Yang, Jinhua; Liu, Huan; Peng, Jiaqi; Zhang, Lijuan

2017-01-01

An adaptive optics (AO) system provides real-time compensation for atmospheric turbulence. However, an AO image is usually of poor contrast because of the nature of the imaging process, meaning that the image contains information coming from both out-of-focus and in-focus planes of the object, which also brings about a loss in quality. In this paper, we present a robust multi-frame adaptive optics image restoration algorithm via maximum likelihood estimation. Our proposed algorithm uses a maximum likelihood method with image regularization as the basic principle, and constructs the joint log likelihood function for multi-frame AO images based on a Poisson distribution model. To begin with, a frame selection method based on image variance is applied to the observed multi-frame AO images to select images with better quality to improve the convergence of a blind deconvolution algorithm. Then, by combining the imaging conditions and the AO system properties, a point spread function estimation model is built. Finally, we develop our iterative solutions for AO image restoration addressing the joint deconvolution issue. We conduct a number of experiments to evaluate the performances of our proposed algorithm. Experimental results show that our algorithm produces accurate AO image restoration results and outperforms the current state-of-the-art blind deconvolution methods. PMID:28383503

GRID-BASED EXPLORATION OF COSMOLOGICAL PARAMETER SPACE WITH SNAKE

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

Mikkelsen, K.; Næss, S. K.; Eriksen, H. K., E-mail: kristin.mikkelsen@astro.uio.no

2013-11-10

We present a fully parallelized grid-based parameter estimation algorithm for investigating multidimensional likelihoods called Snake, and apply it to cosmological parameter estimation. The basic idea is to map out the likelihood grid-cell by grid-cell according to decreasing likelihood, and stop when a certain threshold has been reached. This approach improves vastly on the 'curse of dimensionality' problem plaguing standard grid-based parameter estimation simply by disregarding grid cells with negligible likelihood. The main advantages of this method compared to standard Metropolis-Hastings Markov Chain Monte Carlo methods include (1) trivial extraction of arbitrary conditional distributions; (2) direct access to Bayesian evidences; (3)more » better sampling of the tails of the distribution; and (4) nearly perfect parallelization scaling. The main disadvantage is, as in the case of brute-force grid-based evaluation, a dependency on the number of parameters, N{sub par}. One of the main goals of the present paper is to determine how large N{sub par} can be, while still maintaining reasonable computational efficiency; we find that N{sub par} = 12 is well within the capabilities of the method. The performance of the code is tested by comparing cosmological parameters estimated using Snake and the WMAP-7 data with those obtained using CosmoMC, the current standard code in the field. We find fully consistent results, with similar computational expenses, but shorter wall time due to the perfect parallelization scheme.« less

Estimating Model Probabilities using Thermodynamic Markov Chain Monte Carlo Methods

NASA Astrophysics Data System (ADS)

Ye, M.; Liu, P.; Beerli, P.; Lu, D.; Hill, M. C.

2014-12-01

Markov chain Monte Carlo (MCMC) methods are widely used to evaluate model probability for quantifying model uncertainty. In a general procedure, MCMC simulations are first conducted for each individual model, and MCMC parameter samples are then used to approximate marginal likelihood of the model by calculating the geometric mean of the joint likelihood of the model and its parameters. It has been found the method of evaluating geometric mean suffers from the numerical problem of low convergence rate. A simple test case shows that even millions of MCMC samples are insufficient to yield accurate estimation of the marginal likelihood. To resolve this problem, a thermodynamic method is used to have multiple MCMC runs with different values of a heating coefficient between zero and one. When the heating coefficient is zero, the MCMC run is equivalent to a random walk MC in the prior parameter space; when the heating coefficient is one, the MCMC run is the conventional one. For a simple case with analytical form of the marginal likelihood, the thermodynamic method yields more accurate estimate than the method of using geometric mean. This is also demonstrated for a case of groundwater modeling with consideration of four alternative models postulated based on different conceptualization of a confining layer. This groundwater example shows that model probabilities estimated using the thermodynamic method are more reasonable than those obtained using the geometric method. The thermodynamic method is general, and can be used for a wide range of environmental problem for model uncertainty quantification.

Proactive Control: Neural Oscillatory Correlates of Conflict Anticipation and Response Slowing

PubMed Central

Ide, Jaime S.

2017-01-01

Abstract Proactive control allows us to anticipate environmental changes and adjust behavioral strategy. In the laboratory, investigators have used a number of different behavioral paradigms, including the stop-signal task (SST), to examine the neural processes of proactive control. Previous functional MRI studies of the SST have demonstrated regional responses to conflict anticipation—the likelihood of a stop signal or P(stop) as estimated by a Bayesian model—and reaction time (RT) slowing and how these responses are interrelated. Here, in an electrophysiological study, we investigated the time–frequency domain substrates of proactive control. The results showed that conflict anticipation as indexed by P(stop) was positively correlated with the power in low-theta band (3–5 Hz) in the fixation (trial onset)-locked interval, and go-RT was negatively correlated with the power in delta-theta band (2–8 Hz) in the go-locked interval. Stimulus prediction error was positively correlated with the power in the low-beta band (12–22 Hz) in the stop-locked interval. Further, the power of the P(stop) and go-RT clusters was negatively correlated, providing a mechanism relating conflict anticipation to RT slowing in the SST. Source reconstruction with beamformer localized these time–frequency activities close to brain regions as revealed by functional MRI in earlier work. These are the novel results to show oscillatory electrophysiological substrates in support of trial-by-trial behavioral adjustment for proactive control. PMID:28560315

Risk Indicators for Periodontitis in US Adults: NHANES 2009 to 2012.

PubMed

Eke, Paul I; Wei, Liang; Thornton-Evans, Gina O; Borrell, Luisa N; Borgnakke, Wenche S; Dye, Bruce; Genco, Robert J

2016-10-01

Through the use of optimal surveillance measures and standard case definitions, it is now possible to more accurately determine population-average risk profiles for severe (SP) and non-severe periodontitis (NSP) in adults (aged 30 years and older) in the United States. Data from the 2009 to 2012 National Health and Nutrition Examination Survey were used, which, for the first time, used the "gold standard" full-mouth periodontitis surveillance protocol to classify severity of periodontitis following suggested Centers for Disease Control/American Academy of Periodontology case definitions. Probabilities of periodontitis by: 1) sociodemographics, 2) behavioral factors, and 3) comorbid conditions were assessed using prevalence ratios (PRs) estimated by predicted marginal probability from multivariable generalized logistic regression models. Analyses were further stratified by sex for each classification of periodontitis. Likelihood of total periodontitis (TP) increased with age for overall and NSP relative to non-periodontitis. Compared with non-Hispanic whites, TP was more likely in Hispanics (adjusted [a]PR = 1.38; 95% confidence interval 95% CI: 1.26 to 1.52) and non-Hispanic blacks (aPR = 1.35; 95% CI: 1.22 to 1.50), whereas SP was most likely in non-Hispanic blacks (aPR = 1.82; 95% CI: 1.44 to 2.31). There was at least a 50% greater likelihood of TP in current smokers compared with non-smokers. In males, likelihood of TP in adults aged 65 years and older was greater (aPR = 2.07; 95% CI: 1.76 to 2.43) than adults aged 30 to 44 years. This probability was even greater in women (aPR = 3.15; 95% CI: 2.63 to 3.77). Likelihood of TP was higher in current smokers relative to non-smokers regardless of sex and periodontitis classification. TP was more likely in men with uncontrolled diabetes mellitus (DM) compared with adults without DM. Assessment of risk profiles for periodontitis in adults in the United States based on gold standard periodontal measures show important differences by severity of disease and sex. Cigarette smoking, specifically current smoking, remains an important modifiable risk for all levels of periodontitis severity. Higher likelihood of TP in older adults and in males with uncontrolled DM is noteworthy. These findings could improve identification of target populations for effective public health interventions to improve periodontal health of adults in the United States.

Maximum Likelihood Estimation of Nonlinear Structural Equation Models with Ignorable Missing Data

ERIC Educational Resources Information Center

Lee, Sik-Yum; Song, Xin-Yuan; Lee, John C. K.

2003-01-01

The existing maximum likelihood theory and its computer software in structural equation modeling are established on the basis of linear relationships among latent variables with fully observed data. However, in social and behavioral sciences, nonlinear relationships among the latent variables are important for establishing more meaningful models…

A Composite Likelihood Inference in Latent Variable Models for Ordinal Longitudinal Responses

ERIC Educational Resources Information Center

Vasdekis, Vassilis G. S.; Cagnone, Silvia; Moustaki, Irini

2012-01-01

The paper proposes a composite likelihood estimation approach that uses bivariate instead of multivariate marginal probabilities for ordinal longitudinal responses using a latent variable model. The model considers time-dependent latent variables and item-specific random effects to be accountable for the interdependencies of the multivariate…

Multilevel and Latent Variable Modeling with Composite Links and Exploded Likelihoods

ERIC Educational Resources Information Center

Rabe-Hesketh, Sophia; Skrondal, Anders

2007-01-01

Composite links and exploded likelihoods are powerful yet simple tools for specifying a wide range of latent variable models. Applications considered include survival or duration models, models for rankings, small area estimation with census information, models for ordinal responses, item response models with guessing, randomized response models,…

Estimating residual fault hitting rates by recapture sampling

NASA Technical Reports Server (NTRS)

Lee, Larry; Gupta, Rajan

1988-01-01

For the recapture debugging design introduced by Nayak (1988) the problem of estimating the hitting rates of the faults remaining in the system is considered. In the context of a conditional likelihood, moment estimators are derived and are shown to be asymptotically normal and fully efficient. Fixed sample properties of the moment estimators are compared, through simulation, with those of the conditional maximum likelihood estimators. Properties of the conditional model are investigated such as the asymptotic distribution of linear functions of the fault hitting frequencies and a representation of the full data vector in terms of a sequence of independent random vectors. It is assumed that the residual hitting rates follow a log linear rate model and that the testing process is truncated when the gaps between the detection of new errors exceed a fixed amount of time.

An Example of an Improvable Rao-Blackwell Improvement, Inefficient Maximum Likelihood Estimator, and Unbiased Generalized Bayes Estimator.

PubMed

Galili, Tal; Meilijson, Isaac

2016-01-02

The Rao-Blackwell theorem offers a procedure for converting a crude unbiased estimator of a parameter θ into a "better" one, in fact unique and optimal if the improvement is based on a minimal sufficient statistic that is complete. In contrast, behind every minimal sufficient statistic that is not complete, there is an improvable Rao-Blackwell improvement. This is illustrated via a simple example based on the uniform distribution, in which a rather natural Rao-Blackwell improvement is uniformly improvable. Furthermore, in this example the maximum likelihood estimator is inefficient, and an unbiased generalized Bayes estimator performs exceptionally well. Counterexamples of this sort can be useful didactic tools for explaining the true nature of a methodology and possible consequences when some of the assumptions are violated. [Received December 2014. Revised September 2015.].

New robust statistical procedures for the polytomous logistic regression models.

PubMed

Castilla, Elena; Ghosh, Abhik; Martin, Nirian; Pardo, Leandro

2018-05-17

This article derives a new family of estimators, namely the minimum density power divergence estimators, as a robust generalization of the maximum likelihood estimator for the polytomous logistic regression model. Based on these estimators, a family of Wald-type test statistics for linear hypotheses is introduced. Robustness properties of both the proposed estimators and the test statistics are theoretically studied through the classical influence function analysis. Appropriate real life examples are presented to justify the requirement of suitable robust statistical procedures in place of the likelihood based inference for the polytomous logistic regression model. The validity of the theoretical results established in the article are further confirmed empirically through suitable simulation studies. Finally, an approach for the data-driven selection of the robustness tuning parameter is proposed with empirical justifications. © 2018, The International Biometric Society.

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

PubMed Central

Rosenblum, Michael; van der Laan, Mark J.

2010-01-01

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

ReplacementMatrix: a web server for maximum-likelihood estimation of amino acid replacement rate matrices.

PubMed

Dang, Cuong Cao; Lefort, Vincent; Le, Vinh Sy; Le, Quang Si; Gascuel, Olivier

2011-10-01

Amino acid replacement rate matrices are an essential basis of protein studies (e.g. in phylogenetics and alignment). A number of general purpose matrices have been proposed (e.g. JTT, WAG, LG) since the seminal work of Margaret Dayhoff and co-workers. However, it has been shown that matrices specific to certain protein groups (e.g. mitochondrial) or life domains (e.g. viruses) differ significantly from general average matrices, and thus perform better when applied to the data to which they are dedicated. This Web server implements the maximum-likelihood estimation procedure that was used to estimate LG, and provides a number of tools and facilities. Users upload a set of multiple protein alignments from their domain of interest and receive the resulting matrix by email, along with statistics and comparisons with other matrices. A non-parametric bootstrap is performed optionally to assess the variability of replacement rate estimates. Maximum-likelihood trees, inferred using the estimated rate matrix, are also computed optionally for each input alignment. Finely tuned procedures and up-to-date ML software (PhyML 3.0, XRATE) are combined to perform all these heavy calculations on our clusters. http://www.atgc-montpellier.fr/ReplacementMatrix/ olivier.gascuel@lirmm.fr Supplementary data are available at http://www.atgc-montpellier.fr/ReplacementMatrix/

Inverse problems-based maximum likelihood estimation of ground reflectivity for selected regions of interest from stripmap SAR data [Regularized maximum likelihood estimation of ground reflectivity from stripmap SAR data

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

West, R. Derek; Gunther, Jacob H.; Moon, Todd K.

In this study, we derive a comprehensive forward model for the data collected by stripmap synthetic aperture radar (SAR) that is linear in the ground reflectivity parameters. It is also shown that if the noise model is additive, then the forward model fits into the linear statistical model framework, and the ground reflectivity parameters can be estimated by statistical methods. We derive the maximum likelihood (ML) estimates for the ground reflectivity parameters in the case of additive white Gaussian noise. Furthermore, we show that obtaining the ML estimates of the ground reflectivity requires two steps. The first step amounts tomore » a cross-correlation of the data with a model of the data acquisition parameters, and it is shown that this step has essentially the same processing as the so-called convolution back-projection algorithm. The second step is a complete system inversion that is capable of mitigating the sidelobes of the spatially variant impulse responses remaining after the correlation processing. We also state the Cramer-Rao lower bound (CRLB) for the ML ground reflectivity estimates.We show that the CRLB is linked to the SAR system parameters, the flight path of the SAR sensor, and the image reconstruction grid.We demonstrate the ML image formation and the CRLB bound for synthetically generated data.« less

Inverse problems-based maximum likelihood estimation of ground reflectivity for selected regions of interest from stripmap SAR data [Regularized maximum likelihood estimation of ground reflectivity from stripmap SAR data

DOE PAGES

West, R. Derek; Gunther, Jacob H.; Moon, Todd K.

2016-12-01

In this study, we derive a comprehensive forward model for the data collected by stripmap synthetic aperture radar (SAR) that is linear in the ground reflectivity parameters. It is also shown that if the noise model is additive, then the forward model fits into the linear statistical model framework, and the ground reflectivity parameters can be estimated by statistical methods. We derive the maximum likelihood (ML) estimates for the ground reflectivity parameters in the case of additive white Gaussian noise. Furthermore, we show that obtaining the ML estimates of the ground reflectivity requires two steps. The first step amounts tomore » a cross-correlation of the data with a model of the data acquisition parameters, and it is shown that this step has essentially the same processing as the so-called convolution back-projection algorithm. The second step is a complete system inversion that is capable of mitigating the sidelobes of the spatially variant impulse responses remaining after the correlation processing. We also state the Cramer-Rao lower bound (CRLB) for the ML ground reflectivity estimates.We show that the CRLB is linked to the SAR system parameters, the flight path of the SAR sensor, and the image reconstruction grid.We demonstrate the ML image formation and the CRLB bound for synthetically generated data.« less

Fast Component Pursuit for Large-Scale Inverse Covariance Estimation.

PubMed

Han, Lei; Zhang, Yu; Zhang, Tong

2016-08-01

The maximum likelihood estimation (MLE) for the Gaussian graphical model, which is also known as the inverse covariance estimation problem, has gained increasing interest recently. Most existing works assume that inverse covariance estimators contain sparse structure and then construct models with the ℓ 1 regularization. In this paper, different from existing works, we study the inverse covariance estimation problem from another perspective by efficiently modeling the low-rank structure in the inverse covariance, which is assumed to be a combination of a low-rank part and a diagonal matrix. One motivation for this assumption is that the low-rank structure is common in many applications including the climate and financial analysis, and another one is that such assumption can reduce the computational complexity when computing its inverse. Specifically, we propose an efficient COmponent Pursuit (COP) method to obtain the low-rank part, where each component can be sparse. For optimization, the COP method greedily learns a rank-one component in each iteration by maximizing the log-likelihood. Moreover, the COP algorithm enjoys several appealing properties including the existence of an efficient solution in each iteration and the theoretical guarantee on the convergence of this greedy approach. Experiments on large-scale synthetic and real-world datasets including thousands of millions variables show that the COP method is faster than the state-of-the-art techniques for the inverse covariance estimation problem when achieving comparable log-likelihood on test data.

Frailty Transitions in the San Antonio Longitudinal Study of Aging

PubMed Central

Espinoza, Sara E.; Jung, Inkyung; Hazuda, Helen

2012-01-01

OBJECTIVES To examine frailty transitions in Mexican American (MA) and European American (EA) older adults. DESIGN Longitudinal, observational cohort study. SETTING Socioeconomically diverse neighborhoods in San Antonio, Texas. PARTICIPANTS 312 MA and 285 EA community-dwelling older adults (65+) with frailty information at baseline (1992–96) and transition information at follow-up (2000–01) in the San Antonio Longitudinal Study of Aging (SALSA). MEASUREMENTS Five frailty characteristics (weight loss, exhaustion, weakness, slowness, and low physical activity), frailty score (0–5), and overall frailty state (non-frail = 0 characteristics, pre-frail = 1 or 2, frail = 3+) were assessed at baseline. Transitions (progressed, regressed, or no change) were assessed for frailty score and state. Odds ratios (OR) of progression and regression in individual characteristics were estimated using generalized estimating equations, adjusting for age, sex, ethnic group, socioeconomic status, comorbidity, diabetes, and follow-up interval. RESULTS Diabetes with macrovascular complications (OR=1.84, 95%CI: 1.02–3.33), fewer years of education (OR=0.96, 95%CI: 0.93–1.0) and follow-up interval (OR=1.3, 95%CI: 1.17–1.46) were significant predictors of progression in any frailty characteristic. Mortality increased by frailty state, and pre-frail individuals were more likely than frail to regress. CONCLUSION Diabetes with macrovascular complications and fewer years of education are important predictors of progression in any frailty characteristic. Because of increased risk of death compared with the non-frail state and the increased likelihood of regression compared with the frail state, the pre-frail state may be an optimal target for intervention. PMID:22316162

Mucin Expression in Colorectal Cancer (CRC): Systematic Review and Meta-Analysis.

PubMed

Niv, Yaron; Rokkas, Theodore

2018-05-18

A body of evidence has suggested that mucins play an important role in adhesion, invasion, and cancer metastasis. However, this evidence is scarce and sometimes confusing. We performed a systematic review and meta-analysis of available studies to better define the role of mucins in the behavior of colorectal cancer (CRC). Medical literature was searched through November 30, 2017, using suitable keywords. Pooled estimates, that is, odd ratios (ORs), were obtained using fixed or random-effects models, as appropriate. Heterogeneity between studies was evaluated with the Cochran Q test and I values, whereas the likelihood of publication bias was assessed by constructing funnel plots. Their symmetry was estimated by the Begg and Mazumdar adjusted rank correlation test and by the Egger regression test. A total of 2234 CRC patients were included in 12 studies, eligible for meta-analysis. There was a significant difference concerning total mucin expression between CRC patients and controls [pooled ORs (95% confidence interval)=8.156 (2.624-25.354), test for overall effect Z=3.627, P<0.0001]. There was no significant publication bias. This significant difference was constricting to MUC1. In addition, there was a significance concerning MUC1 overexpression according to the stage of CRC, that is advanced stage versus localized disease [ORs (95% confidence interval)=2.724 (1.211-6.127), Z= 2.423, P=0.015], as opposed to MUC2 and MUC4. MUC1 is overexpressed in CRC tissue comparing with healthy mucosa, and may have a role in the neoplastic transformation and metastatic process. MUC2 has probably no role in carcinogenesis.

The Fecal Microbiota Profile and Bronchiolitis in Infants

PubMed Central

Linnemann, Rachel W.; Mansbach, Jonathan M.; Ajami, Nadim J.; Espinola, Janice A.; Petrosino, Joseph F.; Piedra, Pedro A.; Stevenson, Michelle D.; Sullivan, Ashley F.; Thompson, Amy D.; Camargo, Carlos A.

2016-01-01

BACKGROUND: Little is known about the association of gut microbiota, a potentially modifiable factor, with bronchiolitis in infants. We aimed to determine the association of fecal microbiota with bronchiolitis in infants. METHODS: We conducted a case–control study. As a part of multicenter prospective study, we collected stool samples from 40 infants hospitalized with bronchiolitis. We concurrently enrolled 115 age-matched healthy controls. By applying 16S rRNA gene sequencing and an unbiased clustering approach to these 155 fecal samples, we identified microbiota profiles and determined the association of microbiota profiles with likelihood of bronchiolitis. RESULTS: Overall, the median age was 3 months, 55% were male, and 54% were non-Hispanic white. Unbiased clustering of fecal microbiota identified 4 distinct profiles: Escherichia-dominant profile (30%), Bifidobacterium-dominant profile (21%), Enterobacter/Veillonella-dominant profile (22%), and Bacteroides-dominant profile (28%). The proportion of bronchiolitis was lowest in infants with the Enterobacter/Veillonella-dominant profile (15%) and highest in the Bacteroides-dominant profile (44%), corresponding to an odds ratio of 4.59 (95% confidence interval, 1.58–15.5; P = .008). In the multivariable model, the significant association between the Bacteroides-dominant profile and a greater likelihood of bronchiolitis persisted (odds ratio for comparison with the Enterobacter/Veillonella-dominant profile, 4.24; 95% confidence interval, 1.56–12.0; P = .005). In contrast, the likelihood of bronchiolitis in infants with the Escherichia-dominant or Bifidobacterium-dominant profile was not significantly different compared with those with the Enterobacter/Veillonella-dominant profile. CONCLUSIONS: In this case–control study, we identified 4 distinct fecal microbiota profiles in infants. The Bacteroides-dominant profile was associated with a higher likelihood of bronchiolitis. PMID:27354456

A general methodology for maximum likelihood inference from band-recovery data

USGS Publications Warehouse

Conroy, M.J.; Williams, B.K.

1984-01-01

A numerical procedure is described for obtaining maximum likelihood estimates and associated maximum likelihood inference from band- recovery data. The method is used to illustrate previously developed one-age-class band-recovery models, and is extended to new models, including the analysis with a covariate for survival rates and variable-time-period recovery models. Extensions to R-age-class band- recovery, mark-recapture models, and twice-yearly marking are discussed. A FORTRAN program provides computations for these models.

Association of Air Pollution Exposures With High-Density Lipoprotein Cholesterol and Particle Number: The Multi-Ethnic Study of Atherosclerosis.

PubMed

Bell, Griffith; Mora, Samia; Greenland, Philip; Tsai, Michael; Gill, Ed; Kaufman, Joel D

2017-05-01

The relationship between air pollution and cardiovascular disease may be explained by changes in high-density lipoprotein (HDL). We examined the cross-sectional relationship between air pollution and both HDL cholesterol and HDL particle number in the MESA Air study (Multi-Ethnic Study of Atherosclerosis Air Pollution). Study participants were 6654 white, black, Hispanic, and Chinese men and women aged 45 to 84 years. We estimated individual residential ambient fine particulate pollution exposure (PM 2.5 ) and black carbon concentrations using a fine-scale likelihood-based spatiotemporal model and cohort-specific monitoring. Exposure periods were averaged to 12 months, 3 months, and 2 weeks prior to examination. HDL cholesterol and HDL particle number were measured in the year 2000 using the cholesterol oxidase method and nuclear magnetic resonance spectroscopy, respectively. We used multivariable linear regression to examine the relationship between air pollution exposure and HDL measures. A 0.7×10 - 6 m - 1 higher exposure to black carbon (a marker of traffic-related pollution) averaged over a 1-year period was significantly associated with a lower HDL cholesterol (-1.68 mg/dL; 95% confidence interval, -2.86 to -0.50) and approached significance with HDL particle number (-0.55 mg/dL; 95% confidence interval, -1.13 to 0.03). In the 3-month averaging time period, a 5 μg/m 3 higher PM 2.5 was associated with lower HDL particle number (-0.64 μmol/L; 95% confidence interval, -1.01 to -0.26), but not HDL cholesterol (-0.05 mg/dL; 95% confidence interval, -0.82 to 0.71). These data are consistent with the hypothesis that exposure to air pollution is adversely associated with measures of HDL. © 2017 American Heart Association, Inc.

Meta-Analysis and Systematic Review to Assess the Role of Soluble FMS-Like Tyrosine Kinase-1 and Placenta Growth Factor Ratio in Prediction of Preeclampsia: The SaPPPhirE Study.

PubMed

Agrawal, Swati; Cerdeira, Ana Sofia; Redman, Christopher; Vatish, Manu

2018-02-01

Preeclampsia is a major cause of morbidity and mortality worldwide. Numerous candidate biomarkers have been proposed for diagnosis and prediction of preeclampsia. Measurement of maternal circulating angiogenesis biomarker as the ratio of sFlt-1 (soluble FMS-like tyrosine kinase-1; an antiangiogenic factor)/PlGF (placental growth factor; an angiogenic factor) reflects the antiangiogenic balance that characterizes incipient or overt preeclampsia. The ratio increases before the onset of the disease and thus may help in predicting preeclampsia. We conducted a meta-analysis to explore the predictive accuracy of sFlt-1/PlGF ratio in preeclampsia. We included 15 studies with 534 cases with preeclampsia and 19 587 controls. The ratio has a pooled sensitivity of 80% (95% confidence interval, 0.68-0.88), specificity of 92% (95% confidence interval, 0.87-0.96), positive likelihood ratio of 10.5 (95% confidence interval, 6.2-18.0), and a negative likelihood ratio of 0.22 (95% confidence interval, 0.13-0.35) in predicting preeclampsia in both high- and low-risk patients. Most of the studies have not made a distinction between early- and late-onset disease, and therefore, the analysis for it could not be done. It can prove to be a valuable screening tool for preeclampsia and may also help in decision-making, treatment stratification, and better resource allocation. © 2017 American Heart Association, Inc.

Likelihood-based gene annotations for gap filling and quality assessment in genome-scale metabolic models

DOE PAGES

Benedict, Matthew N.; Mundy, Michael B.; Henry, Christopher S.; ...

2014-10-16

Genome-scale metabolic models provide a powerful means to harness information from genomes to deepen biological insights. With exponentially increasing sequencing capacity, there is an enormous need for automated reconstruction techniques that can provide more accurate models in a short time frame. Current methods for automated metabolic network reconstruction rely on gene and reaction annotations to build draft metabolic networks and algorithms to fill gaps in these networks. However, automated reconstruction is hampered by database inconsistencies, incorrect annotations, and gap filling largely without considering genomic information. Here we develop an approach for applying genomic information to predict alternative functions for genesmore » and estimate their likelihoods from sequence homology. We show that computed likelihood values were significantly higher for annotations found in manually curated metabolic networks than those that were not. We then apply these alternative functional predictions to estimate reaction likelihoods, which are used in a new gap filling approach called likelihood-based gap filling to predict more genomically consistent solutions. To validate the likelihood-based gap filling approach, we applied it to models where essential pathways were removed, finding that likelihood-based gap filling identified more biologically relevant solutions than parsimony-based gap filling approaches. We also demonstrate that models gap filled using likelihood-based gap filling provide greater coverage and genomic consistency with metabolic gene functions compared to parsimony-based approaches. Interestingly, despite these findings, we found that likelihoods did not significantly affect consistency of gap filled models with Biolog and knockout lethality data. This indicates that the phenotype data alone cannot necessarily be used to discriminate between alternative solutions for gap filling and therefore, that the use of other information is necessary to obtain a more accurate network. All described workflows are implemented as part of the DOE Systems Biology Knowledgebase (KBase) and are publicly available via API or command-line web interface.« less

Likelihood-Based Gene Annotations for Gap Filling and Quality Assessment in Genome-Scale Metabolic Models

PubMed Central

Benedict, Matthew N.; Mundy, Michael B.; Henry, Christopher S.; Chia, Nicholas; Price, Nathan D.

2014-01-01

Genome-scale metabolic models provide a powerful means to harness information from genomes to deepen biological insights. With exponentially increasing sequencing capacity, there is an enormous need for automated reconstruction techniques that can provide more accurate models in a short time frame. Current methods for automated metabolic network reconstruction rely on gene and reaction annotations to build draft metabolic networks and algorithms to fill gaps in these networks. However, automated reconstruction is hampered by database inconsistencies, incorrect annotations, and gap filling largely without considering genomic information. Here we develop an approach for applying genomic information to predict alternative functions for genes and estimate their likelihoods from sequence homology. We show that computed likelihood values were significantly higher for annotations found in manually curated metabolic networks than those that were not. We then apply these alternative functional predictions to estimate reaction likelihoods, which are used in a new gap filling approach called likelihood-based gap filling to predict more genomically consistent solutions. To validate the likelihood-based gap filling approach, we applied it to models where essential pathways were removed, finding that likelihood-based gap filling identified more biologically relevant solutions than parsimony-based gap filling approaches. We also demonstrate that models gap filled using likelihood-based gap filling provide greater coverage and genomic consistency with metabolic gene functions compared to parsimony-based approaches. Interestingly, despite these findings, we found that likelihoods did not significantly affect consistency of gap filled models with Biolog and knockout lethality data. This indicates that the phenotype data alone cannot necessarily be used to discriminate between alternative solutions for gap filling and therefore, that the use of other information is necessary to obtain a more accurate network. All described workflows are implemented as part of the DOE Systems Biology Knowledgebase (KBase) and are publicly available via API or command-line web interface. PMID:25329157

Free kick instead of cross-validation in maximum-likelihood refinement of macromolecular crystal structures

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

Pražnikar, Jure; University of Primorska,; Turk, Dušan, E-mail: dusan.turk@ijs.si

2014-12-01

The maximum-likelihood free-kick target, which calculates model error estimates from the work set and a randomly displaced model, proved superior in the accuracy and consistency of refinement of crystal structures compared with the maximum-likelihood cross-validation target, which calculates error estimates from the test set and the unperturbed model. The refinement of a molecular model is a computational procedure by which the atomic model is fitted to the diffraction data. The commonly used target in the refinement of macromolecular structures is the maximum-likelihood (ML) function, which relies on the assessment of model errors. The current ML functions rely on cross-validation. Theymore » utilize phase-error estimates that are calculated from a small fraction of diffraction data, called the test set, that are not used to fit the model. An approach has been developed that uses the work set to calculate the phase-error estimates in the ML refinement from simulating the model errors via the random displacement of atomic coordinates. It is called ML free-kick refinement as it uses the ML formulation of the target function and is based on the idea of freeing the model from the model bias imposed by the chemical energy restraints used in refinement. This approach for the calculation of error estimates is superior to the cross-validation approach: it reduces the phase error and increases the accuracy of molecular models, is more robust, provides clearer maps and may use a smaller portion of data for the test set for the calculation of R{sub free} or may leave it out completely.« less