<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet type="text/xsl" href="static/style.xsl"?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-22T09:28:10Z</responseDate><request verb="GetRecord" identifier="oai:uwspace.uwaterloo.ca:10012/4041" metadataPrefix="dim">https://dspacemainprd01.lib.uwaterloo.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:uwspace.uwaterloo.ca:10012/4041</identifier><datestamp>2017-03-21T16:14:24Z</datestamp><setSpec>com_10012_9924</setSpec><setSpec>com_10012_1</setSpec><setSpec>col_10012_9934</setSpec><setSpec>col_10012_6</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:doc="http://www.lyncode.com/xoai" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="author" authority="9122d796-c30d-44ef-ad87-0cded06770f5" confidence="600">Chen, Baojiang</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2008-09-25T18:21:24Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="available">2008-09-25T18:21:24Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2008-09-25T18:21:24Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="submitted">2008</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">http://hdl.handle.net/10012/4041</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract" lang="en">Analyses of longitudinal categorical data are typically based on&#xd;
semiparametric models in which covariate effects are expressed on&#xd;
marginal probabilities and estimation is carried out based on&#xd;
generalized estimating equations (GEE). Methods based on GEE are&#xd;
motivated in part by the lack of tractable models for clustered&#xd;
categorical data. However such marginal methods may not yield fully&#xd;
efficient estimates, nor consistent estimates when missing data are&#xd;
present. In the first part of the thesis I develop a Markov model&#xd;
for the analysis of longitudinal categorical data which facilitates&#xd;
modeling marginal and conditional structures. A likelihood&#xd;
formulation is employed for inference, so the resulting estimators&#xd;
enjoy properties such as optimal efficiency and consistency, and&#xd;
remain consistent when data are missing at random. Simulation&#xd;
studies demonstrate that the proposed method performs well under a&#xd;
variety of situations. Application to data from a smoking prevention&#xd;
study illustrates the utility of the model and interpretation of&#xd;
covariate effects.&#xd;
&#xd;
Incomplete data often arise in many areas of research in practice.&#xd;
 This phenomenon is&#xd;
common in longitudinal data on disease history of subjects.&#xd;
Progressive models provide a convenient framework for characterizing&#xd;
disease processes which arise, for example, when the state&#xd;
represents the degree of the irreversible damage incurred by the&#xd;
subject. Problems arise if the mechanism leading to the missing data&#xd;
is related to the response process. A naive analysis might lead to&#xd;
biased results and invalid inferences. The second part of this&#xd;
thesis begins with an investigation of progressive multi-state&#xd;
models for longitudinal studies with incomplete observations.&#xd;
Maximum likelihood estimation is carried out based on an EM&#xd;
algorithm, and variance estimation is provided using Louis method.&#xd;
In general, the maximum likelihood estimates are valid when the&#xd;
missing data mechanism is missing completely at random or missing at&#xd;
random. Here we provide likelihood based method in that the&#xd;
parameters are identifiable no matter what the missing data&#xd;
mechanism. Simulation studies demonstrate that the proposed method&#xd;
works well under a variety of situations.&#xd;
&#xd;
In practice, we often face data with missing values in both the&#xd;
response and the covariates, and sometimes there is some association&#xd;
between the missingness of the response and the covariate. The&#xd;
proper analysis of this type of data requires taking this&#xd;
correlation into consideration. The impact of attrition in&#xd;
longitudinal studies depends on the correlation between the missing&#xd;
response and missing covariate. Ignoring such correlation can bias&#xd;
the statistical inference. We have studied the proper method that&#xd;
incorporates the association between the missingness of the response&#xd;
and missing covariate through the use of inverse probability&#xd;
weighted generalized estimating equations. The simulation&#xd;
illustrates that the proposed method yields a consistent estimator,&#xd;
while the method that ignores the association yields an inconsistent&#xd;
estimator.&#xd;
&#xd;
Many analyses for longitudinal incomplete data focus on studying the&#xd;
impact of covariates on the mean responses. However, little&#xd;
attention has been directed to address the impact of missing&#xd;
covariates on the association parameters in clustered longitudinal&#xd;
studies. The last part of this thesis mainly addresses this problem.&#xd;
Weighted first and second order estimating equations are constructed&#xd;
to obtain consistent estimates of mean and association parameters.</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso" lang="en">en</dim:field>
   <dim:field mdschema="dc" element="publisher" lang="en">University of Waterloo</dim:field>
   <dim:field mdschema="dc" element="title" lang="en">Statistical Methods for Multi-State Analysis of Incomplete Longitudinal Data</dim:field>
   <dim:field mdschema="dc" element="type" lang="en">Doctoral Thesis</dim:field>
   <dim:field mdschema="dc" element="pending" lang="en">false</dim:field>
   <dim:field mdschema="dc" element="subject" qualifier="program" lang="en">Statistics (Biostatistics)</dim:field>
   <dim:field mdschema="uws-etd" element="degree" qualifier="department" lang="en">Statistics and Actuarial Science</dim:field>
   <dim:field mdschema="uws-etd" element="degree" lang="en">Doctor of Philosophy</dim:field>
   <dim:field mdschema="uws" element="typeOfResource" lang="en">Text</dim:field>
   <dim:field mdschema="uws" element="peerReviewStatus" lang="en">Unreviewed</dim:field>
   <dim:field mdschema="uws" element="scholarLevel" lang="en">Graduate</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
</dim:dim>
</metadata></record></GetRecord></OAI-PMH>