<?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>2026-11-22T15:46:50Z</responseDate><request verb="GetRecord" identifier="oai:uwspace.uwaterloo.ca:10012/4614" metadataPrefix="dim">https://dspacemainprd01.lib.uwaterloo.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:uwspace.uwaterloo.ca:10012/4614</identifier><datestamp>2017-03-21T16:47:23Z</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="79b0ed42-d629-428e-8334-df26d3de03df" confidence="600">Zhou, Qian</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2009-08-26T20:41:50Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="available">2009-08-26T20:41:50Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2009-08-26T20:41:50Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="submitted">2009</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">http://hdl.handle.net/10012/4614</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract" lang="en">Estimating functions have been widely used for parameter&#xd;
estimation in various statistical problems. Regular estimating&#xd;
functions produce parameter estimators which have desirable&#xd;
properties, such as consistency and asymptotic normality. In&#xd;
quasi-likelihood inference, an important example of estimating&#xd;
functions, correct specification of the first two moments of the&#xd;
underlying distribution leads to the information unbiasedness, which&#xd;
states that two forms of the information matrix: the negative&#xd;
sensitivity matrix (negative expectation of the first order&#xd;
derivative of an estimating function) and the variability matrix&#xd;
(variance of an estimating function) are equal, or in other words,&#xd;
the analogue of the Fisher information is equivalent to the Godambe&#xd;
information. Consequently, the information unbiasedness indicates&#xd;
that the model-based covariance matrix estimator and sandwich&#xd;
covariance matrix estimator are equivalent. By comparing the&#xd;
model-based and sandwich variance estimators, we propose information&#xd;
ratio (IR) statistics for testing model misspecification of&#xd;
variance/covariance structure under correctly specified mean&#xd;
structure, in the context of linear regression models, generalized&#xd;
linear regression models and generalized estimating equations.&#xd;
Asymptotic properties of the IR statistics are discussed. In&#xd;
addition, through intensive simulation studies, we show that the IR&#xd;
statistics are powerful in various applications: test for&#xd;
heteroscedasticity in linear regression models, test for&#xd;
overdispersion in count data, and test for misspecified variance&#xd;
function and/or misspecified working correlation structure.&#xd;
Moreover, the IR statistics appear more powerful than the classical&#xd;
information matrix test proposed by White (1982).&#xd;
&#xd;
In the literature, model selection criteria have been intensively&#xd;
discussed, but almost all of them target choosing the optimal mean&#xd;
structure. In this thesis, two model selection procedures are&#xd;
proposed for selecting the optimal variance/covariance structure&#xd;
among a collection of candidate structures. One is based on a&#xd;
sequence of the IR tests for all the competing variance/covariance&#xd;
structures. The other is based on an ``information discrepancy&#xd;
criterion&amp;quot; (IDC), which provides a measurement of discrepancy&#xd;
between the negative sensitivity matrix and the variability matrix.&#xd;
In fact, this IDC characterizes the relative efficiency loss when&#xd;
using a certain candidate variance/covariance structure, compared&#xd;
with the true but unknown structure. Through simulation studies and&#xd;
analyses of two data sets, it is shown that the two proposed model&#xd;
selection methods both have a high rate of detecting the&#xd;
true/optimal variance/covariance structure. In particular, since the&#xd;
IDC magnifies the difference among the competing structures, it is&#xd;
highly sensitive to detect the most appropriate variance/covariance&#xd;
structure.</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="subject" lang="en">Estimating functions</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Information unbiasedness</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Quasi-likelihood inference</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Model misspecfication</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Model selection</dim:field>
   <dim:field mdschema="dc" element="subject" qualifier="program" lang="en">Statistics (Biostatistics)</dim:field>
   <dim:field mdschema="dc" element="title" lang="en">Information Matrices in Estimating Function Approach: Tests for Model Misspecification and Model Selection</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="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>