Statistics and Probability Commons^™

Linear Mixed-Effects Models: Applications To The Behavioral Sciences And Adolescent Community Health, Lizmarie Gabriela Maldonado

USF Tampa Graduate Theses and Dissertations

Linear mixed-effects (LME) modeling is a widely used statistical method for analyzing repeated measures or longitudinal data. Such longitudinal studies typically aim to investigate and describe the trajectory of a desired outcome. Longitudinal data have the advantage over cross-sectional data by providing more accuracy for the model. LME models allow researchers to account for random variation among individuals and between individuals.

In this project, adolescent health was chosen as a topic of research due to the many changes that occur during this crucial time period as a precursor to overall well-being in adult life. Understanding the factors that influence how …

Higher Order Markov Structure-Based Logistic Model And Likelihood Inference For Ordinal Data, Soma Chowdhury Biswas, M. Ataharul Islam, Jamal Nazrul Islam

Journal of Modern Applied Statistical Methods

Azzalini (1994) proposed a first order Markov chain for binary data. Azzalini’s model is extended for ordinal data and introduces a second order model. Further, the test statistics are developed and the power of the test is determined. An application using real data is also presented.

Longitudinal Analysis Of Spatiotemporal Processes: A Case Study Of Dynamic Contrast-Enhanced Magnetic Resonance Imaging In Multiple Sclerosis, Russell T. Shinohara, Ciprian M. Crainiceanu, Brian S. Caffo, Daniel S. Reich

Johns Hopkins University, Dept. of Biostatistics Working Papers

Multiple sclerosis (MS) is an immune-mediated disease in which inflammatory lesions form in the brain. In many active MS lesions, the blood-brain barrier (BBB) is disrupted and blood flows into white matter; this disruption may be related to morbidity and disability. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) allows quantitative study of blood flow and permeability dynamics throughout the brain. This technique involves a subject being imaged sequentially during a study visit as an intravenously administered contrast agent flows into the brain. In regions where flow is abnormal, such as white matter lesions, this allows the quantification of the BBB damage. …

Bayesian Semiparametric Generalizations Of Linear Models Using Polya Trees, Angela Schoergendorfer

University of Kentucky Doctoral Dissertations

In a Bayesian framework, prior distributions on a space of nonparametric continuous distributions may be defined using Polya trees. This dissertation addresses statistical problems for which the Polya tree idea can be utilized to provide efficient and practical methodological solutions.

One problem considered is the estimation of risks, odds ratios, or other similar measures that are derived by specifying a threshold for an observed continuous variable. It has been previously shown that fitting a linear model to the continuous outcome under the assumption of a logistic error distribution leads to more efficient odds ratio estimates. We will show that deviations …

Reducing Selection Bias In Analyzing Longitudinal Health Data With High Mortality Rates, Xian Liu, Charles C. Engel, Han Kang, Kristie L. Gore

Journal of Modern Applied Statistical Methods

Two longitudinal regression models, one parametric and one nonparametric, are developed to reduce selection bias when analyzing longitudinal health data with high mortality rates. The parametric mixed model is a two-step linear regression approach, whereas the nonparametric mixed-effects regression model uses a retransformation method to handle random errors across time.

Canonical Correlation Analysis For Longitudinal Data, Raymond Mccollum

Mathematics & Statistics Theses & Dissertations

Data (multivariate data) on two sets of vectors commonly occur in applications. Statistical analysis of these data is usually done using a canonical correlation analysis (CCA). Occurrence of these data at multiple occasions or conditions leads to longitudinal multivariate data for a CCA. We address the problem of canonical correlation analysis on longitudinal data when the data have a Kronecker product covariance structure. Using structured correlation matrices we model the dependency of repeatedly observed data. Recent work of Srivastava, Nahtman, and von Rosen (2008) developed an iterative algorithm to determine the maximum likelihood estimate of the Kronecker product covariance structure …

Marginal Regression Modeling Under Irregular, Biased Sampling, Petra Buzkova, Thomas Lumley

UW Biostatistics Working Paper Series

In longitudinal studies observations are often obtained at continuous subject-specific times. Frequently the availability of outcome data may be related to the outcome measure or other covariates that are related to the outcome measure. Under such biased sampling designs unadjusted regression analysis yield biased estimates. Building on the work of Lin & Ying (2001) that integrates counting processes techniques with longitudinal data settings we propose a class of estimators that can handle biased sampling. We call those estimators ``inverse--intensity--rate--ratio--weighted'' (IIRR) estimators. Of major focus is a mean--response model where we examine the marginal effect of the covariate X at time …

Longitudinal Data Analysis For Generalized Linear Models Under Irregular, Biased Sampling: Situations With Follow-Up Dependent On Outcome Or Auxiliary Outcome-Related Variables, Petra Buzkova, Thomas Lumley

UW Biostatistics Working Paper Series

In longitudinal studies, observations are often obtained at subject-specific observation times. Those times can be continuous times, not at a set of prespecified times. Frequently the observation times may be related to the outcome measure or other auxiliary variables that are related to the outcome measure but undesirable to condition upon in the regression model for outcome. Regression analysis unadjusted for such sampling designs yield biased estimates. Based on estimating equations, we propose a class of estimators in generalized linear regression models that can handle biased sampling under continuous observation times. We call those estimators ``inverse--intensity rate--ratio--weighted'' (IIRR) estimators. The …

Semiparametric Loglinear Regression For Longitudinal Measurements Subject To Irregular, Biased Follow-Up, Petra Buzkova, Thomas Lumley

UW Biostatistics Working Paper Series

We propose a method for analysis of loglinear regression models for longitudinal data that are subject to continuous and irregular follow-up. Frequently, if the follow-up is irregular, the availability of outcome data may be related to the outcome measure or other covariates that are related to the outcome measure. Under such biased sampling designs unadjusted regression analysis yield biased estimates. We examine the marginal association of the covariates X at time t and the logarithm of the mean of response Y at time t. We focus on semiparametric regression with unspecified baseline function of time. To predict the follow-up times …

G-Computation Estimation Of Nonparametric Causal Effects On Time-Dependent Mean Outcomes In Longitudinal Studies, Romain Neugebauer, Mark J. Van Der Laan

U.C. Berkeley Division of Biostatistics Working Paper Series

Two approaches to Causal Inference based on Marginal Structural Models (MSM) have been proposed. They provide different representations of causal effects with distinct causal parameters. Initially, a parametric MSM approach to Causal Inference was developed: it relies on correct specification of a parametric MSM. Recently, a new approach based on nonparametric MSM was introduced. This later approach does not require the assumption of a correctly specified MSM and thus is more realistic if one believes that correct specification of a parametric MSM is unlikely in practice. However, this approach was described only for investigating causal effects on mean outcomes collected …

Statistical Analysis Of Longitudinal And Multivariate Discrete Data, Deepak Mav

Mathematics & Statistics Theses & Dissertations

Correlated multivariate Poisson and binary variables occur naturally in medical, biological and epidemiological longitudinal studies. Modeling and simulating such variables is difficult because the correlations are restricted by the marginal means via Fréchet bounds in a complicated way. In this dissertation we will first discuss partially specified models and methods for estimating the regression and correlation parameters. We derive the asymptotic distributions of these parameter estimates. Using simulations based on extensions of the algorithm due to Sim (1993, Journal of Statistical Computation and Simulation, 47, pp. 1–10), we study the performance of these estimates using infeasibility, coverage probabilities of the …

Modeling Incomplete Longitudinal Data, Hakan Demirtas

Journal of Modern Applied Statistical Methods

This article presents a review of popular parametric, semiparametric and ad-hoc approaches for analyzing incomplete longitudinal data.

History-Adjusted Marginal Structural Models And Statically-Optimal Dynamic Treatment Regimes, Mark J. Van Der Laan, Maya L. Petersen

U.C. Berkeley Division of Biostatistics Working Paper Series

Marginal structural models (MSM) provide a powerful tool for estimating the causal effect of a treatment. These models, introduced by Robins, model the marginal distributions of treatment-specific counterfactual outcomes, possibly conditional on a subset of the baseline covariates. Marginal structural models are particularly useful in the context of longitudinal data structures, in which each subject's treatment and covariate history are measured over time, and an outcome is recorded at a final time point. However, the utility of these models for some applications has been limited by their inability to incorporate modification of the causal effect of treatment by time-varying covariates. …

Estimation Of Direct And Indirect Causal Effects In Longitudinal Studies, Mark J. Van Der Laan, Maya L. Petersen

U.C. Berkeley Division of Biostatistics Working Paper Series

The causal effect of a treatment on an outcome is generally mediated by several intermediate variables. Estimation of the component of the causal effect of a treatment that is mediated by a given intermediate variable (the indirect effect of the treatment), and the component that is not mediated by that intermediate variable (the direct effect of the treatment) is often relevant to mechanistic understanding and to the design of clinical and public health interventions. Under the assumption of no-unmeasured confounders, Robins & Greenland (1992) and Pearl (2000), develop two identifiability results for direct and indirect causal effects. They define an …

Marginal Modeling Of Multilevel Binary Data With Time-Varying Covariates, Diana Miglioretti, Patrick Heagerty

UW Biostatistics Working Paper Series

We propose and compare two approaches for regression analysis of multilevel binary data when clusters are not necessarily nested: a GEE method that relies on a working independence assumption coupled with a three-step method for obtaining empirical standard errors; and a likelihood-based method implemented using Bayesian computational techniques. Implications of time-varying endogenous covariates are addressed. The methods are illustrated using data from the Breast Cancer Surveillance Consortium to estimate mammography accuracy from a repeatedly screened population.

Comparison Of Viral Trajectories In Aids Studies By Using Nonparametric Mixed-Effects Models, Chin-Shang Li, Hua Liang, Ying-Hen Hsieh, Shiing-Jer Twu

Journal of Modern Applied Statistical Methods

The efficacy of antiretroviral therapies for human immunodeficiency virus (HIV) infection can be assessed by studying the trajectory of the changing viral load with treatment time, but estimation of viral trajectory parameters by using the implicit function form of linear and nonlinear parametric models can be problematic. Using longitudinal viral load data from a clinical study of HIV-infected patients in Taiwan, we described the viral trajectories by applying a nonparametric mixed-effects model. We were then able to compare the efficacies of highly active antiretroviral therapy (HAART) and conventional therapy by using Young and Bowman’s (1995) test.

Efficient Semiparametric Marginal Estimation For Longitudinal/Clustered Data, Naisyin Wang, Raymond J. Carroll, Xihong Lin

The University of Michigan Department of Biostatistics Working Paper Series

We consider marginal generalized semiparametric partially linear models for clustered data. Lin and Carroll (2001a) derived the semiparametric efficinet score funtion for this problem in the mulitvariate Gaussian case, but they were unable to contruct a semiparametric efficient estimator that actually achieved the semiparametric information bound. We propose such an estimator here and generalize the work to marginal generalized partially liner models. Asymptotic relative efficincies of the estimation or throughout are investigated. The finite sample performance of these estimators is evaluated through simulations and illustrated using a longtiudinal CD4 count data set. Both theoretical and numerical results indicate that properly …

Equivalent Kernels Of Smoothing Splines In Nonparametric Regression For Clustered/Longitudinal Data, Xihong Lin, Naisyin Wang, Alan H. Welsh, Raymond J. Carroll

The University of Michigan Department of Biostatistics Working Paper Series

We compare spline and kernel methods for clustered/longitudinal data. For independent data, it is well known that kernel methods and spline methods are essentially asymptotically equivalent (Silverman, 1984). However, the recent work of Welsh, et al. (2002) shows that the same is not true for clustered/longitudinal data. First, conventional kernel methods fail to account for the within- cluster correlation, while spline methods are able to account for this correlation. Second, kernel methods and spline methods were found to have different local behavior, with conventional kernels being local and splines being non-local. To resolve these differences, we show that a smoothing …

Histospline Method In Nonparametric Regression Models With Application To Clustered/Longitudinal Data, Raymond J. Carroll, Peter Hall, Tatiyana V. Apanasovich, Xihong Lin

The University of Michigan Department of Biostatistics Working Paper Series

Kernel and smoothing methods for nonparametric function and curve estimation have been particularly successful in "standard" settings, where function values are observed subject to independent errors. However, when aspects of the function are known parametrically, or where the sampling scheme has significant structure, it can be quite difficult to adapt standard methods in such a way that they retain good statistical performance and continue to enjoy easy computability and good numerical properties. In particular, when using local linear modeling it is often awkward to both respect the sampling scheme and produce an estimator with good variance properties, without resorting to …

Double Robust Estimation In Longitudinal Marginal Structural Models, Zhuo Yu, Mark J. Van Der Laan

U.C. Berkeley Division of Biostatistics Working Paper Series

Consider estimation of causal parameters in a marginal structural model for the discrete intensity of the treatment specific counting process (e.g. hazard of a treatment specific survival time) based on longitudinal observational data on treatment, covariates and survival. We assume the sequential randomization assumption (SRA) on the treatment assignment mechanism and the so called experimental treatment assignment assumption which is needed to identify the causal parameters from the observed data distribution. Under SRA, the likelihood of the observed data structure factorizes in the auxiliary treatment mechanism and the partial likelihood consisting of the product over time of conditional distributions of …

Chronic Disease Data And Analysis: Current State Of The Field, Ralph D'Agostino Sr., Lisa M. Sullivan

Journal of Modern Applied Statistical Methods

Chronic disease usually spans years of a person’s lifetime and includes a disease free period, a preclinical, or latent period, where there are few overt signs of disease, a clinical period where the disease manifests and is eventually diagnosed, and a follow-up period where the disease might progress steadily or remain stable. It is often of interest to investigate the relationship between risk factors measured at a point in time (usually during the disease free or preclinical period), and the development of disease at some future point (e.g., 10 years later). We outline some popular designs for the identification of …

A Longitudinal Follow-Up Of Discrete Mass At Zero With Gap, Joseph L. Musial, Patrick D. Bridge, Nicol R. Shamey

Journal of Modern Applied Statistical Methods

The first part of this paper discusses a five-year systematic review of the Journal of Consulting and Clinical Psychology following the landmark power study conducted by Sawilowsky and Hillman (1992). The second part discusses a five-year longitudinal follow-up of a radically nonnormal population distribution: discrete mass at zero with gap. This distribution was based upon a real dataset.