Modification of the generalized quasi-likelihood model in the analysis of the Add Health study

Katherine E. Irimata, Jeffrey R. Wilson

Research output: Contribution to journalArticle

Abstract

The relationship between the mean and variance is an implicit assumption of parametric modeling. While many distributions in the exponential family have a theoretical mean–variance relationship, it is often the case that the data under investigation are correlated, thus varying from the relation. We present a generalized method of moments estimation technique for modeling certain correlated data by adjusting the mean–variance relationship parameters based on a canonical parameterization. The proposed mean–variance form describes overdispersion using two parameters and implements an adjusted canonical parameter which makes this approach feasible for all distributions in the exponential family. Test statistics and confidence intervals are used to measure the deviations from the mean–variance relation parameters. We use the modified relation as a means of fitting generalized quasi-likelihood models to correlated data. The performance of the proposed modified generalized quasi-likelihood model is demonstrated through a simulation study and the importance of accounting for overdispersion is highlighted through the evaluation of adolescent obesity data collected from a U.S. longitudinal study.

Original languageEnglish (US)
JournalStatistical Methods in Medical Research
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

National Longitudinal Study of Adolescent Health
Quasi-likelihood
Overdispersion
Correlated Data
Health
Exponential Family
Pediatric Obesity
Moment Estimation
Longitudinal Studies
Parametric Modeling
Generalized Method of Moments
Obesity
Longitudinal Study
Confidence Intervals
Parameterization
Test Statistic
Confidence interval
Two Parameters
Deviation
Simulation Study

Keywords

  • Canonical parameter
  • correlation
  • generalized linear models
  • generalized method of moments
  • overdispersion

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability
  • Health Information Management

Cite this

@article{646bc9ae832f40cf8e58771c7ae2817b,
title = "Modification of the generalized quasi-likelihood model in the analysis of the Add Health study",
abstract = "The relationship between the mean and variance is an implicit assumption of parametric modeling. While many distributions in the exponential family have a theoretical mean–variance relationship, it is often the case that the data under investigation are correlated, thus varying from the relation. We present a generalized method of moments estimation technique for modeling certain correlated data by adjusting the mean–variance relationship parameters based on a canonical parameterization. The proposed mean–variance form describes overdispersion using two parameters and implements an adjusted canonical parameter which makes this approach feasible for all distributions in the exponential family. Test statistics and confidence intervals are used to measure the deviations from the mean–variance relation parameters. We use the modified relation as a means of fitting generalized quasi-likelihood models to correlated data. The performance of the proposed modified generalized quasi-likelihood model is demonstrated through a simulation study and the importance of accounting for overdispersion is highlighted through the evaluation of adolescent obesity data collected from a U.S. longitudinal study.",
keywords = "Canonical parameter, correlation, generalized linear models, generalized method of moments, overdispersion",
author = "Irimata, {Katherine E.} and Wilson, {Jeffrey R.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1177/0962280219884980",
language = "English (US)",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",

}

TY - JOUR

T1 - Modification of the generalized quasi-likelihood model in the analysis of the Add Health study

AU - Irimata, Katherine E.

AU - Wilson, Jeffrey R.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - The relationship between the mean and variance is an implicit assumption of parametric modeling. While many distributions in the exponential family have a theoretical mean–variance relationship, it is often the case that the data under investigation are correlated, thus varying from the relation. We present a generalized method of moments estimation technique for modeling certain correlated data by adjusting the mean–variance relationship parameters based on a canonical parameterization. The proposed mean–variance form describes overdispersion using two parameters and implements an adjusted canonical parameter which makes this approach feasible for all distributions in the exponential family. Test statistics and confidence intervals are used to measure the deviations from the mean–variance relation parameters. We use the modified relation as a means of fitting generalized quasi-likelihood models to correlated data. The performance of the proposed modified generalized quasi-likelihood model is demonstrated through a simulation study and the importance of accounting for overdispersion is highlighted through the evaluation of adolescent obesity data collected from a U.S. longitudinal study.

AB - The relationship between the mean and variance is an implicit assumption of parametric modeling. While many distributions in the exponential family have a theoretical mean–variance relationship, it is often the case that the data under investigation are correlated, thus varying from the relation. We present a generalized method of moments estimation technique for modeling certain correlated data by adjusting the mean–variance relationship parameters based on a canonical parameterization. The proposed mean–variance form describes overdispersion using two parameters and implements an adjusted canonical parameter which makes this approach feasible for all distributions in the exponential family. Test statistics and confidence intervals are used to measure the deviations from the mean–variance relation parameters. We use the modified relation as a means of fitting generalized quasi-likelihood models to correlated data. The performance of the proposed modified generalized quasi-likelihood model is demonstrated through a simulation study and the importance of accounting for overdispersion is highlighted through the evaluation of adolescent obesity data collected from a U.S. longitudinal study.

KW - Canonical parameter

KW - correlation

KW - generalized linear models

KW - generalized method of moments

KW - overdispersion

UR - http://www.scopus.com/inward/record.url?scp=85074832299&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85074832299&partnerID=8YFLogxK

U2 - 10.1177/0962280219884980

DO - 10.1177/0962280219884980

M3 - Article

C2 - 31686601

AN - SCOPUS:85074832299

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

ER -