@article { ISI:000354249800006, title = {A flexible mixed-effect negative binomial regression model for detecting unusual increases in MRI lesion counts in individual multiple sclerosis patients}, journal = {Statistics in Medicine}, volume = {34}, number = {13}, year = {2015}, pages = {2165-2180}, publisher = {WILEY-BLACKWELL}, address = {111 RIVER ST, HOBOKEN 07030-5774, NJ USA}, abstract = {

We develop a new modeling approach to enhance a recently proposed method to detect increases of contrast-enhancing lesions (CELs) on repeated magnetic resonance imaging, which have been used as an indicator for potential adverse events in multiple sclerosis clinical trials. The method signals patients with unusual increases in CEL activity by estimating the probability of observing CEL counts as large as those observed on a patient{\textquoteright}s recent scans conditional on the patient{\textquoteright}s CEL counts on previous scans. This conditional probability index (CPI), computed based on a mixed-effect negative binomial regression model, can vary substantially depending on the choice of distribution for the patient-specific random effects. Therefore, we relax this parametric assumption to model the random effects with an infinite mixture of beta distributions, using the Dirichlet process, which effectively allows any form of distribution. To our knowledge, no previous literature considers a mixed-effect regression for longitudinal count variables where the random effect is modeled with a Dirichlet process mixture. As our inference is in the Bayesian framework, we adopt a meta-analytic approach to develop an informative prior based on previous clinical trials. This is particularly helpful at the early stages of trials when less data are available. Our enhanced method is illustrated with CEL data from 10 previous multiple sclerosis clinical trials. Our simulation study shows that our procedure estimates the CPI more accurately than parametric alternatives when the patient-specific random effect distribution is misspecified and that an informative prior improves the accuracy of the CPI estimates. Copyright (c) 2015 John Wiley \& Sons, Ltd.

}, keywords = {Dirichlet process, Longitudinal data, meta-analysis, Negative binomial, nonparametric Bayesian procedures, safety monitoring in clinical trials}, doi = {10.1002/sim.6484}, author = {Kondo, Y and Zhao, Y and Petkau, J} } @article {qiu_moving_2015, title = {A moving blocks empirical likelihood method for longitudinal data}, journal = {Biometrics}, volume = {71}, number = {3}, year = {2015}, month = {sep}, pages = {616{\textendash}624}, abstract = {In the analysis of longitudinal or panel data, neglecting the serial correlations among the repeated measurements within subjects may lead to inefficient inference. In particular, when the number of repeated measurements is large, it may be desirable to model the serial correlations more generally. An appealing approach is to accommodate the serial correlations nonparametrically. In this article, we propose a moving blocks empirical likelihood method for general estimating equations. Asymptotic results are derived under sequential limits. Simulation studies are conducted to investigate the finite sample performances of the proposed methods and compare them with the elementwise and subject-wise empirical likelihood methods of Wang et al. (2010, Biometrika 97, 79{\textendash}93) and the block empirical likelihood method of You et al. (2006, Can. J. Statist. 34, 79{\textendash}96). An application to an AIDS longitudinal study is presented.}, keywords = {Empirical likelihood, General estimating equation, Longitudinal data, Nonparametric method, Serial correlation}, issn = {1541-0420}, doi = {10.1111/biom.12317}, url = {http://onlinelibrary.wiley.com/doi/10.1111/biom.12317/abstract}, author = {Qiu, Jin and WU, LANG} } @article { ISI:000309793400023, title = {Pair copula constructions for multivariate discrete data}, journal = {Journal of the American Statistical Association}, volume = {107}, number = {499}, year = {2012}, month = {SEP}, pages = {1063-1072}, publisher = {Amer Statistical Assoc}, type = {Article}, abstract = {Multivariate discrete response data can be found in diverse fields, including econometrics, finance, biometrics, and psychometrics. Our contribution, through this study, is to introduce a new class of models for multivariate discrete data based on pair copula constructions (PCCs) that has two major advantages. First, by deriving the conditions under which any multivariate discrete distribution can be decomposed as a PCC, we show that discrete PCCs attain highly flexible dependence structures. Second, the computational burden of evaluating the likelihood for an m-dimensional discrete PCC only grows quadratically with in. This compares favorably to existing models for which computing the likelihood either requires the evaluation of 2(m) terms or slow numerical integration methods. We demonstrate the high quality of inference function for margins and maximum likelihood estimates, both under a simulated setting and for an application to a longitudinal discrete dataset on headache severity. This article has online supplementary material.}, keywords = {D-vine, Inference function for margins, Longitudinal data, Model selection, Ordered probit regression}, issn = {0162-1459}, doi = {10.1080/01621459.2012.682850}, author = {Panagiotelis, Anastasios and Czado, Claudia and Joe, Harry} } @article {huang_bayesian_2011, title = {Bayesian inference on joint models of HIV dynamics for time-to-event and longitudinal data with skewness and covariate measurement errors}, journal = {Statistics in Medicine}, volume = {30}, number = {24}, year = {2011}, pages = {2930{\textendash}2946}, abstract = {Normality (symmetry) of the model random errors is a routine assumption for mixed-effects models in many longitudinal studies, but it may be unrealistically obscuring important features of subject variations. Covariates are usually introduced in the models to partially explain inter-subject variations, but some covariates such as CD4 cell count may be often measured with substantial errors. This paper formulates a class of models in general forms that considers model errors to have skew-normal distributions for a joint behavior of longitudinal dynamic processes and time-to-event process of interest. For estimating model parameters, we propose a Bayesian approach to jointly model three components (response, covariate, and time-to-event processes) linked through the random effects that characterize the underlying individual-specific longitudinal processes. We discuss in detail special cases of the model class, which are offered to jointly model HIV dynamic response in the presence of CD4 covariate process with measurement errors and time to decrease in CD4/CD8 ratio, to provide a tool to assess antiretroviral treatment and to monitor disease progression. We illustrate the proposed methods using the data from a clinical trial study of HIV treatment. The findings from this research suggest that the joint models with a skew-normal distribution may provide more reliable and robust results if the data exhibit skewness, and particularly the results may be important for HIV/AIDS studies in providing quantitative guidance to better understand the virologic responses to antiretroviral treatment. Copyright {\textcopyright} 2011 John Wiley \& Sons, Ltd.}, keywords = {Bayesian analysis, covariate measurement errors, joint mixed-effects models, Longitudinal data, skew-normal distribution, time to event}, issn = {1097-0258}, doi = {10.1002/sim.4321}, url = {http://onlinelibrary.wiley.com/doi/10.1002/sim.4321/abstract}, author = {Huang, Yangxin and Dagne, Getachew and WU, LANG} } @article {yi_simultaneous_2011, title = {Simultaneous Inference and Bias Analysis for Longitudinal Data with Covariate Measurement Error and Missing Responses}, journal = {Biometrics}, volume = {67}, number = {1}, year = {2011}, month = {mar}, pages = {67{\textendash}75}, abstract = {Summary Longitudinal data arise frequently in medical studies and it is common practice to analyze such data with generalized linear mixed models. Such models enable us to account for various types of heterogeneity, including between- and within-subjects ones. Inferential procedures complicate dramatically when missing observations or measurement error arise. In the literature, there has been considerable interest in accommodating either incompleteness or covariate measurement error under random effects models. However, there is relatively little work concerning both features simultaneously. There is a need to fill up this gap as longitudinal data do often have both characteristics. In this article, our objectives are to study simultaneous impact of missingness and covariate measurement error on inferential procedures and to develop a valid method that is both computationally feasible and theoretically valid. Simulation studies are conducted to assess the performance of the proposed method, and a real example is analyzed with the proposed method.}, keywords = {Bias analysis, Longitudinal data, Measurement error, missing data, Monte Carlo EM algorithm, Random effects models}, issn = {1541-0420}, doi = {10.1111/j.1541-0420.2010.01437.x}, url = {http://onlinelibrary.wiley.com/doi/10.1111/j.1541-0420.2010.01437.x/abstract}, author = {Yi, G. Y. and Liu, W. and WU, LANG} } @article { ISI:000276284600019, title = {Generating random AR(p) and MA(q) Toeplitz correlation matrices}, journal = {Journal of Multivariate Analysis}, volume = {101}, number = {6}, year = {2010}, month = {JUL}, pages = {1532-1545}, publisher = {Elsevier Inc}, type = {Article}, abstract = {Methods are proposed for generating random (p+1) x (p+1) Toeplitz correlation matrices that are consistent with a causal AR(p) Gaussian time series model. The main idea is to first specify distributions for the partial autocorrelations that are algebraically independent and take values in (-1, 1), and then map to the Toeplitz matrix. Similarly, starting with pseudopartial autocorrelations, methods are proposed for generating (q+1) x (q+1) Toeplitz correlation matrices that are consistent with an invertible MA(q) Gaussian time series model. The density can be uniform or non-uniform over the space of autocorrelations up to lag p or q, or over the space of autoregressive or moving average coefficients, by making appropriate choices for the densities of the (pseudo)-partial autocorrelations. Important intermediate steps are the derivations of the Jacobians of the mappings between the (pseudo)-partial autocorrelations, autocorrelations and autoregressive/moving average coefficients. The random generating methods are useful for models with a structured Toeplitz matrix as a parameter. (C) 2010 Elsevier Inc. All rights reserved.}, keywords = {Autoregressive process, Beta distribution, Longitudinal data, Moving average process}, issn = {0047-259X}, doi = {10.1016/j.jmva.2010.01.013}, author = {Ng, C. T. and Joe, Harry} } @article {wu_joint_2010, title = {Joint Inference on HIV Viral Dynamics and Immune Suppression in Presence of Measurement Errors}, journal = {Biometrics}, volume = {66}, number = {2}, year = {2010}, month = {jun}, pages = {327{\textendash}335}, abstract = {Summary: In an attempt to provide a tool to assess antiretroviral therapy and to monitor disease progression, this article studies association of human immunodeficiency virus (HIV) viral suppression and immune restoration. The data from a recent acquired immune deficiency syndrome (AIDS) study are used for illustration. We jointly model HIV viral dynamics and time to decrease in CD4/CD8 ratio in the presence of CD4 process with measurement errors, and estimate the model parameters simultaneously via a method based on a Laplace approximation and the commonly used Monte Carlo EM algorithm. The approaches and many of the points presented apply generally.}, keywords = {Laplace approximation, Longitudinal data, Mixed-effects, Nonlinear models, Time-to-event}, issn = {1541-0420}, doi = {10.1111/j.1541-0420.2009.01308.x}, url = {http://onlinelibrary.wiley.com/doi/10.1111/j.1541-0420.2009.01308.x/abstract}, author = {Wu, L. and Liu, W. and Hu, X. J.} } @article {liu_asymptotic_2010, title = {Some asymptotic results for semiparametric nonlinear mixed-effects models with incomplete data}, journal = {Journal of Statistical Planning and Inference}, volume = {140}, number = {1}, year = {2010}, month = {jan}, pages = {52{\textendash}64}, abstract = {In modeling complex longitudinal data, semiparametric nonlinear mixed-effects (SNLME) models are very flexible and useful. Covariates are often introduced in the models to partially explain the inter-individual variations. In practice, data are often incomplete in the sense that there are often measurement errors and missing data in longitudinal studies. The likelihood method is a standard approach for inference for these models but it can be computationally very challenging, so computationally efficient approximate methods are quite valuable. However, the performance of these approximate methods is often based on limited simulation studies, and theoretical results are unavailable for many approximate methods. In this article, we consider a computationally efficient approximate method for a class of SNLME models with incomplete data and investigate its theoretical properties. We show that the estimates based on the approximate method are consistent and asymptotically normally distributed.}, keywords = {Approximation, Asymptotics, Longitudinal data, Measurement error}, issn = {0378-3758}, doi = {10.1016/j.jspi.2009.06.006}, url = {http://www.sciencedirect.com/science/article/pii/S0378375809001888}, author = {Liu, Wei and WU, LANG} } @article { ISI:000263136700009, title = {On weighting of bivariate margins in pairwise likelihood}, journal = {Journal of Multivariate Analysis}, volume = {100}, number = {4}, year = {2009}, month = {APR}, pages = {670-685}, publisher = {Elsevier Inc}, type = {article}, abstract = {Composite and pairwise likelihood methods have recently been increasingly used. For clustered data with varying cluster sizes, we study asymptotic relative efficiencies for various weighted pairwise likelihoods, with weight being a function of cluster size. For longitudinal data, we also study weighted pairwise likelihoods with weights that can depend on lag. Good choice of weights are needed to avoid the undesirable behavior of estimators with low efficiency. Some analytic results are obtained using the multivariate normal distribution. For clustered data, a practically good choice of weight is obtained after study of relative efficiencies for an exchangeable multivariate normal model; they are different from weights that had previously been suggested. For longitudinal data, there are advantages to only include bivariate margins of adjacent or nearly adjacent pairs in the weighted pairwise likelihood. (C) 2008 Elsevier Inc. All rights reserved.}, keywords = {Binary probit, Clustered data, Composite likelihood, Longitudinal data}, issn = {0047-259X}, doi = {10.1016/j.jmva.2008.07.004}, author = {Joe, Harry and Lee, Youngjo} } @article {wu_approximate_2008, title = {An approximate method for nonlinear mixed-effects models with nonignorably missing covariates}, journal = {Statistics \& Probability Letters}, volume = {78}, number = {4}, year = {2008}, pages = {384{\textendash}389}, abstract = {Nonlinear mixed-effect (NLME) models are very useful in many longitudinal studies. In practice, covariates in NLME models may contain missing data, and the missing data may be nonignorable. Likelihood inference for NLME models with missing covariates can be computationally very intensive. We propose a computationally much more efficient approximate method for NLME models with nonignorably missing covariates. We illustrate the method using a real data example.}, keywords = {EM algorithm, Linearization, Longitudinal data, Taylor expansion}, issn = {0167-7152}, doi = {10.1016/j.spl.2007.07.011}, url = {http://www.sciencedirect.com/science/article/pii/S0167715207002519}, author = {WU, LANG} } @article {wu_joint_2008, title = {Joint inference for nonlinear mixed-effects models and time to event at the presence of missing data}, journal = {Biostatistics}, volume = {9}, number = {2}, year = {2008}, pages = {308{\textendash}320}, abstract = {In many longitudinal studies, the individual characteristics associated with the repeated measures may be possible covariates of the time to an event of interest, and thus, it is desirable to model the time-to-event process and the longitudinal process jointly. Statistical analyses may be further complicated in such studies with missing data such as informative dropouts. This article considers a nonlinear mixed-effects model for the longitudinal process and the Cox proportional hazards model for the time-to-event process. We provide a method for simultaneous likelihood inference on the 2 models and allow for nonignorable data missing. The approach is illustrated with a recent AIDS study by jointly modeling HIV viral dynamics and time to viral rebound.}, keywords = {EM algorithm, Longitudinal data, proportional hazards model, shared parameter model}, issn = {1465-4644, 1468-4357}, doi = {10.1093/biostatistics/kxm029}, url = {http://biostatistics.oxfordjournals.org/content/9/2/308}, author = {WU, LANG and Hu, X. Joan and Wu, Hulin} } @article {wu_computationally_2007, title = {A computationally efficient method for nonlinear mixed-effects models with nonignorable missing data in time-varying covariates}, journal = {Computational Statistics \& Data Analysis}, volume = {51}, number = {5}, year = {2007}, pages = {2410{\textendash}2419}, abstract = {Nonlinear mixed-effects (NLME) models are widely used for longitudinal data analyses. Time-dependent covariates are often introduced to partially explain inter-individual variation. These covariates often have missing data, and the missingness may be nonignorable. Likelihood inference for NLME models with nonignorable missing data in time-varying covariates can be computationally very intensive and may even offer computational difficulties such as nonconvergence. We propose a computationally very efficient method for approximate likelihood inference. The method is illustrated using a real data example.}, keywords = {EM algorithm, Linearization, Longitudinal data, Random effects model}, issn = {0167-9473}, doi = {10.1016/j.csda.2006.07.036}, url = {http://www.sciencedirect.com/science/article/pii/S0167947306002556}, author = {WU, LANG} } @article {wu_hiv_2007, title = {HIV viral dynamic models with dropouts and missing covariates}, journal = {Statistics in Medicine}, volume = {26}, number = {17}, year = {2007}, pages = {3342{\textendash}3357}, abstract = {In recent years HIV viral dynamic models have received great attention in AIDS studies. Often, subjects in these studies may drop out for various reasons such as drug intolerance or drug resistance, and covariates may also contain missing data. Statistical analyses ignoring informative dropouts and missing covariates may lead to misleading results. We consider appropriate methods for HIV viral dynamic models with informative dropouts and missing covariates and evaluate these methods via simulations. A real data set is analysed, and the results show that the initial viral decay rate, which may reflect the efficacy of the anti-HIV treatment, may be over-estimated if dropout patients are ignored. We also find that the current or immediate previous viral load values may be most predictive for patients{\textquoteright} dropout. These results may be important for HIV/AIDS studies. Copyright {\textcopyright} 2007 John Wiley \& Sons, Ltd.}, keywords = {approximate method, Longitudinal data, missing data, Monte-Carlo EM, nonlinear mixed-effects model}, issn = {1097-0258}, doi = {10.1002/sim.2816}, url = {http://onlinelibrary.wiley.com/doi/10.1002/sim.2816/abstract}, author = {WU, LANG} } @article {liu_simultaneous_2007, title = {Simultaneous Inference for Semiparametric Nonlinear Mixed-Effects Models with Covariate Measurement Errors and Missing Responses}, journal = {Biometrics}, volume = {63}, number = {2}, year = {2007}, month = {jun}, pages = {342{\textendash}350}, abstract = {Summary Semiparametric nonlinear mixed-effects (NLME) models are flexible for modeling complex longitudinal data. Covariates are usually introduced in the models to partially explain interindividual variations. Some covariates, however, may be measured with substantial errors. Moreover, the responses may be missing and the missingness may be nonignorable. We propose two approximate likelihood methods for semiparametric NLME models with covariate measurement errors and nonignorable missing responses. The methods are illustrated in a real data example. Simulation results show that both methods perform well and are much better than the commonly used naive method.}, keywords = {Cubic spline basis, Longitudinal data, Monte Carlo EM algorithm, Random-effects model}, issn = {1541-0420}, doi = {10.1111/j.1541-0420.2006.00687.x}, url = {http://onlinelibrary.wiley.com/doi/10.1111/j.1541-0420.2006.00687.x/abstract}, author = {Liu, Wei and WU, LANG} } @article {wu_nonlinear_2004, title = {Nonlinear mixed-effect models with nonignorably missing covariates}, journal = {Canadian Journal of Statistics}, volume = {32}, number = {1}, year = {2004}, month = {mar}, pages = {27{\textendash}37}, abstract = {Nonlinear mixed-effect models are often used in the analysis of longitudinal data. However, it sometimes happens that missing values for some of the model covariates are not purely random. Motivated by an application to HTV viral dynamics, where this situation occurs, the author considers likelihood inference for this type of problem. His approach involves a Monte Carlo EM algorithm, along with a Gibbs sampler and rejection/importance sampling methods. A concrete application is provided.}, keywords = {EM algorithm, Gibbs sampling, Longitudinal data, missing data, Rejection sampling}, issn = {1708-945X}, doi = {10.2307/3315997}, url = {http://onlinelibrary.wiley.com/doi/10.2307/3315997/abstract}, author = {WU, LANG} }