Partitioned GMM logistic regression models for longitudinal data

Research output: Contribution to journalArticle

Abstract

Correlation is inherent in longitudinal studies due to the repeated measurements on subjects, as well as due to time-dependent covariates in the study. In the National Longitudinal Study of Adolescent to Adult Health (Add Health), data were repeatedly collected on children in grades 7-12 across four waves. Thus, observations obtained on the same adolescent were correlated, while predictors were correlated with current and future outcomes such as obesity status, among other health issues. Previous methods, such as the generalized method of moments (GMM) approach have been proposed to estimate regression coefficients for time-dependent covariates. However, these approaches combined all valid moment conditions to produce an averaged parameter estimate for each covariate and thus assumed that the effect of each covariate on the response was constant across time. This assumption is not necessarily optimal in applications such as Add Health or health-related data. Thus, we depart from this assumption and instead use the Partitioned GMM approach to estimate multiple coefficients for the data based on different time periods. These extra regression coefficients are obtained using a partitioning of the moment conditions pertaining to each respective relationship. This approach offers a deeper understanding and appreciation into the effect of each covariate on the response. We conduct simulation studies, as well as analyses of obesity in Add Health, rehospitalization in Medicare data, and depression scores in a clinical study. The Partitioned GMM methods exhibit benefits over previously proposed models with improved insight into the nonconstant relationships realized when analyzing longitudinal data.

Original languageEnglish (US)
JournalStatistics in Medicine
DOIs
StatePublished - Jan 1 2019

Fingerprint

Generalized Method of Moments
Logistic Regression Model
Longitudinal Data
Health
Logistic Models
Time-dependent Covariates
Covariates
Obesity
Moment Conditions
Longitudinal Study
Regression Coefficient
Longitudinal Studies
Estimate
Repeated Measurements
Moment Method
Pediatric Obesity
Medicare
Time Constant
Predictors
Partitioning

Keywords

  • generalized method of moments
  • logistic regression
  • longitudinal data
  • repeated measures
  • time-dependent covariates

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Partitioned GMM logistic regression models for longitudinal data. / Irimata, Kyle M.; Broatch, Jennifer; Wilson, Jeffrey.

In: Statistics in Medicine, 01.01.2019.

Research output: Contribution to journalArticle

@article{e25327321cfb4ef79e4f8623e053f6d6,
title = "Partitioned GMM logistic regression models for longitudinal data",
abstract = "Correlation is inherent in longitudinal studies due to the repeated measurements on subjects, as well as due to time-dependent covariates in the study. In the National Longitudinal Study of Adolescent to Adult Health (Add Health), data were repeatedly collected on children in grades 7-12 across four waves. Thus, observations obtained on the same adolescent were correlated, while predictors were correlated with current and future outcomes such as obesity status, among other health issues. Previous methods, such as the generalized method of moments (GMM) approach have been proposed to estimate regression coefficients for time-dependent covariates. However, these approaches combined all valid moment conditions to produce an averaged parameter estimate for each covariate and thus assumed that the effect of each covariate on the response was constant across time. This assumption is not necessarily optimal in applications such as Add Health or health-related data. Thus, we depart from this assumption and instead use the Partitioned GMM approach to estimate multiple coefficients for the data based on different time periods. These extra regression coefficients are obtained using a partitioning of the moment conditions pertaining to each respective relationship. This approach offers a deeper understanding and appreciation into the effect of each covariate on the response. We conduct simulation studies, as well as analyses of obesity in Add Health, rehospitalization in Medicare data, and depression scores in a clinical study. The Partitioned GMM methods exhibit benefits over previously proposed models with improved insight into the nonconstant relationships realized when analyzing longitudinal data.",
keywords = "generalized method of moments, logistic regression, longitudinal data, repeated measures, time-dependent covariates",
author = "Irimata, {Kyle M.} and Jennifer Broatch and Jeffrey Wilson",
year = "2019",
month = "1",
day = "1",
doi = "10.1002/sim.8099",
language = "English (US)",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",

}

TY - JOUR

T1 - Partitioned GMM logistic regression models for longitudinal data

AU - Irimata, Kyle M.

AU - Broatch, Jennifer

AU - Wilson, Jeffrey

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Correlation is inherent in longitudinal studies due to the repeated measurements on subjects, as well as due to time-dependent covariates in the study. In the National Longitudinal Study of Adolescent to Adult Health (Add Health), data were repeatedly collected on children in grades 7-12 across four waves. Thus, observations obtained on the same adolescent were correlated, while predictors were correlated with current and future outcomes such as obesity status, among other health issues. Previous methods, such as the generalized method of moments (GMM) approach have been proposed to estimate regression coefficients for time-dependent covariates. However, these approaches combined all valid moment conditions to produce an averaged parameter estimate for each covariate and thus assumed that the effect of each covariate on the response was constant across time. This assumption is not necessarily optimal in applications such as Add Health or health-related data. Thus, we depart from this assumption and instead use the Partitioned GMM approach to estimate multiple coefficients for the data based on different time periods. These extra regression coefficients are obtained using a partitioning of the moment conditions pertaining to each respective relationship. This approach offers a deeper understanding and appreciation into the effect of each covariate on the response. We conduct simulation studies, as well as analyses of obesity in Add Health, rehospitalization in Medicare data, and depression scores in a clinical study. The Partitioned GMM methods exhibit benefits over previously proposed models with improved insight into the nonconstant relationships realized when analyzing longitudinal data.

AB - Correlation is inherent in longitudinal studies due to the repeated measurements on subjects, as well as due to time-dependent covariates in the study. In the National Longitudinal Study of Adolescent to Adult Health (Add Health), data were repeatedly collected on children in grades 7-12 across four waves. Thus, observations obtained on the same adolescent were correlated, while predictors were correlated with current and future outcomes such as obesity status, among other health issues. Previous methods, such as the generalized method of moments (GMM) approach have been proposed to estimate regression coefficients for time-dependent covariates. However, these approaches combined all valid moment conditions to produce an averaged parameter estimate for each covariate and thus assumed that the effect of each covariate on the response was constant across time. This assumption is not necessarily optimal in applications such as Add Health or health-related data. Thus, we depart from this assumption and instead use the Partitioned GMM approach to estimate multiple coefficients for the data based on different time periods. These extra regression coefficients are obtained using a partitioning of the moment conditions pertaining to each respective relationship. This approach offers a deeper understanding and appreciation into the effect of each covariate on the response. We conduct simulation studies, as well as analyses of obesity in Add Health, rehospitalization in Medicare data, and depression scores in a clinical study. The Partitioned GMM methods exhibit benefits over previously proposed models with improved insight into the nonconstant relationships realized when analyzing longitudinal data.

KW - generalized method of moments

KW - logistic regression

KW - longitudinal data

KW - repeated measures

KW - time-dependent covariates

UR - http://www.scopus.com/inward/record.url?scp=85060974257&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060974257&partnerID=8YFLogxK

U2 - 10.1002/sim.8099

DO - 10.1002/sim.8099

M3 - Article

C2 - 30701570

AN - SCOPUS:85060974257

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

ER -