Estimating equations for a latent transition model with multiple discrete indicators

Beth A. Reboussin, Kung Yee Liang, David M. Reboussin

Research output: Contribution to journalArticle

Abstract

This paper proposes a two-part model for studying transitions between health states over time when multiple, discrete health indicators are available. The includes a measurement model positing underlying latent health states and a transition model between latent health states over time. Full maximum likelihood estimation procedures are computationally complex in this latent variable framework, making only a limited class of models feasible and estimation of standard errors problematic. For this reason, an estimating equations analogue of the pseudo-likelihood method for the parameters of interest, namely the transition model parameters, is considered. The finite sample properties of the proposed procedure are investigated through a simulation study and the importance of choosing strong indicators of the latent variable is demonstrated. The applicability of the methodology is illustrated with health survey data measuring disability in the elderly from the Longitudinal Study of Aging.

Original languageEnglish (US)
Pages (from-to)839-845
Number of pages7
JournalBiometrics
Volume55
Issue number3
StatePublished - Sep 1999

Fingerprint

Transition Model
Estimating Equation
Health
Health Transition
Latent Variables
Health Surveys
Longitudinal Studies
Pseudo-likelihood
Likelihood Methods
Longitudinal Study
Maximum likelihood estimation
Survey Data
Disability
Standard error
longitudinal studies
Maximum Likelihood Estimation
Aging of materials
Simulation Study
Model
Analogue

Keywords

  • Estimating equations
  • Latent transition models
  • Multivariate categorical data

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Public Health, Environmental and Occupational Health
  • Agricultural and Biological Sciences (miscellaneous)
  • Applied Mathematics
  • Statistics and Probability

Cite this

Reboussin, B. A., Liang, K. Y., & Reboussin, D. M. (1999). Estimating equations for a latent transition model with multiple discrete indicators. Biometrics, 55(3), 839-845.

Estimating equations for a latent transition model with multiple discrete indicators. / Reboussin, Beth A.; Liang, Kung Yee; Reboussin, David M.

In: Biometrics, Vol. 55, No. 3, 09.1999, p. 839-845.

Research output: Contribution to journalArticle

Reboussin, BA, Liang, KY & Reboussin, DM 1999, 'Estimating equations for a latent transition model with multiple discrete indicators', Biometrics, vol. 55, no. 3, pp. 839-845.
Reboussin BA, Liang KY, Reboussin DM. Estimating equations for a latent transition model with multiple discrete indicators. Biometrics. 1999 Sep;55(3):839-845.
Reboussin, Beth A. ; Liang, Kung Yee ; Reboussin, David M. / Estimating equations for a latent transition model with multiple discrete indicators. In: Biometrics. 1999 ; Vol. 55, No. 3. pp. 839-845.
@article{b069d810ae3441908b91f133d6c27ab0,
title = "Estimating equations for a latent transition model with multiple discrete indicators",
abstract = "This paper proposes a two-part model for studying transitions between health states over time when multiple, discrete health indicators are available. The includes a measurement model positing underlying latent health states and a transition model between latent health states over time. Full maximum likelihood estimation procedures are computationally complex in this latent variable framework, making only a limited class of models feasible and estimation of standard errors problematic. For this reason, an estimating equations analogue of the pseudo-likelihood method for the parameters of interest, namely the transition model parameters, is considered. The finite sample properties of the proposed procedure are investigated through a simulation study and the importance of choosing strong indicators of the latent variable is demonstrated. The applicability of the methodology is illustrated with health survey data measuring disability in the elderly from the Longitudinal Study of Aging.",
keywords = "Estimating equations, Latent transition models, Multivariate categorical data",
author = "Reboussin, {Beth A.} and Liang, {Kung Yee} and Reboussin, {David M.}",
year = "1999",
month = "9",
language = "English (US)",
volume = "55",
pages = "839--845",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - Estimating equations for a latent transition model with multiple discrete indicators

AU - Reboussin, Beth A.

AU - Liang, Kung Yee

AU - Reboussin, David M.

PY - 1999/9

Y1 - 1999/9

N2 - This paper proposes a two-part model for studying transitions between health states over time when multiple, discrete health indicators are available. The includes a measurement model positing underlying latent health states and a transition model between latent health states over time. Full maximum likelihood estimation procedures are computationally complex in this latent variable framework, making only a limited class of models feasible and estimation of standard errors problematic. For this reason, an estimating equations analogue of the pseudo-likelihood method for the parameters of interest, namely the transition model parameters, is considered. The finite sample properties of the proposed procedure are investigated through a simulation study and the importance of choosing strong indicators of the latent variable is demonstrated. The applicability of the methodology is illustrated with health survey data measuring disability in the elderly from the Longitudinal Study of Aging.

AB - This paper proposes a two-part model for studying transitions between health states over time when multiple, discrete health indicators are available. The includes a measurement model positing underlying latent health states and a transition model between latent health states over time. Full maximum likelihood estimation procedures are computationally complex in this latent variable framework, making only a limited class of models feasible and estimation of standard errors problematic. For this reason, an estimating equations analogue of the pseudo-likelihood method for the parameters of interest, namely the transition model parameters, is considered. The finite sample properties of the proposed procedure are investigated through a simulation study and the importance of choosing strong indicators of the latent variable is demonstrated. The applicability of the methodology is illustrated with health survey data measuring disability in the elderly from the Longitudinal Study of Aging.

KW - Estimating equations

KW - Latent transition models

KW - Multivariate categorical data

UR - http://www.scopus.com/inward/record.url?scp=0032887133&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032887133&partnerID=8YFLogxK

M3 - Article

VL - 55

SP - 839

EP - 845

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 3

ER -