A semivarying joint model for longitudinal binary and continuous outcomes

Esra Kürüm, John Hughes, Runze Li

Research output: Contribution to journalArticle

Abstract

Semivarying models extend varying coefficient models by allowing some regression coefficients to be constant with respect to the underlying covariate(s). In this paper we develop a semivarying joint modelling framework for estimating the time-varying association between two intensively measured longitudinal responses: a continuous one and a binary one. To overcome the major challenge of jointly modelling these responses, namely, the lack of a natural multivariate distribution we introduce a Gaussian latent variable underlying the binary response. We then decompose the model into two components: a marginal model for the continuous response and a conditional model for the binary response given the continuous response. We develop a two-stage estimation procedure and discuss the asymptotic normality of the resulting estimators. We assess the finite-sample performance of our procedure using a simulation study, and we illustrate our method by analyzing binary and continuous responses from the Women's Interagency HIV Study.

Original languageEnglish (US)
Pages (from-to)44-57
Number of pages14
JournalCanadian Journal of Statistics
Volume44
Issue number1
DOIs
StatePublished - Mar 1 2016

Fingerprint

Joint Model
Binary
Binary Response
Two-stage Estimation
Marginal Model
Varying Coefficient Model
Joint Modeling
Conditional Model
Multivariate Distribution
Latent Variables
Regression Coefficient
Asymptotic Normality
Covariates
Time-varying
Simulation Study
Estimator
Decompose
Modeling
Model

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{b050fc9c00f546f480e59c80a54853c8,
title = "A semivarying joint model for longitudinal binary and continuous outcomes",
abstract = "Semivarying models extend varying coefficient models by allowing some regression coefficients to be constant with respect to the underlying covariate(s). In this paper we develop a semivarying joint modelling framework for estimating the time-varying association between two intensively measured longitudinal responses: a continuous one and a binary one. To overcome the major challenge of jointly modelling these responses, namely, the lack of a natural multivariate distribution we introduce a Gaussian latent variable underlying the binary response. We then decompose the model into two components: a marginal model for the continuous response and a conditional model for the binary response given the continuous response. We develop a two-stage estimation procedure and discuss the asymptotic normality of the resulting estimators. We assess the finite-sample performance of our procedure using a simulation study, and we illustrate our method by analyzing binary and continuous responses from the Women's Interagency HIV Study.",
author = "Esra K{\"u}r{\"u}m and John Hughes and Runze Li",
year = "2016",
month = "3",
day = "1",
doi = "10.1002/cjs.11273",
language = "English (US)",
volume = "44",
pages = "44--57",
journal = "Canadian Journal of Statistics",
issn = "0319-5724",
publisher = "Statistical Society of Canada",
number = "1",

}

A semivarying joint model for longitudinal binary and continuous outcomes. / Kürüm, Esra; Hughes, John; Li, Runze.

In: Canadian Journal of Statistics, Vol. 44, No. 1, 01.03.2016, p. 44-57.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A semivarying joint model for longitudinal binary and continuous outcomes

AU - Kürüm, Esra

AU - Hughes, John

AU - Li, Runze

PY - 2016/3/1

Y1 - 2016/3/1

N2 - Semivarying models extend varying coefficient models by allowing some regression coefficients to be constant with respect to the underlying covariate(s). In this paper we develop a semivarying joint modelling framework for estimating the time-varying association between two intensively measured longitudinal responses: a continuous one and a binary one. To overcome the major challenge of jointly modelling these responses, namely, the lack of a natural multivariate distribution we introduce a Gaussian latent variable underlying the binary response. We then decompose the model into two components: a marginal model for the continuous response and a conditional model for the binary response given the continuous response. We develop a two-stage estimation procedure and discuss the asymptotic normality of the resulting estimators. We assess the finite-sample performance of our procedure using a simulation study, and we illustrate our method by analyzing binary and continuous responses from the Women's Interagency HIV Study.

AB - Semivarying models extend varying coefficient models by allowing some regression coefficients to be constant with respect to the underlying covariate(s). In this paper we develop a semivarying joint modelling framework for estimating the time-varying association between two intensively measured longitudinal responses: a continuous one and a binary one. To overcome the major challenge of jointly modelling these responses, namely, the lack of a natural multivariate distribution we introduce a Gaussian latent variable underlying the binary response. We then decompose the model into two components: a marginal model for the continuous response and a conditional model for the binary response given the continuous response. We develop a two-stage estimation procedure and discuss the asymptotic normality of the resulting estimators. We assess the finite-sample performance of our procedure using a simulation study, and we illustrate our method by analyzing binary and continuous responses from the Women's Interagency HIV Study.

UR - http://www.scopus.com/inward/record.url?scp=84958861868&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958861868&partnerID=8YFLogxK

U2 - 10.1002/cjs.11273

DO - 10.1002/cjs.11273

M3 - Article

C2 - 27667895

AN - SCOPUS:84958861868

VL - 44

SP - 44

EP - 57

JO - Canadian Journal of Statistics

JF - Canadian Journal of Statistics

SN - 0319-5724

IS - 1

ER -