A Bayesian semiparametric model for bivariate sparse longitudinal data

Kiranmoy Das, Runze Li, Subhajit Sengupta, Rongling Wu

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Mixed-effects models have recently become popular for analyzing sparse longitudinal data that arise naturally in biological, agricultural and biomedical studies. Traditional approaches assume independent residuals over time and explain the longitudinal dependence by random effects. However, when bivariate or multivariate traits are measured longitudinally, this fundamental assumption is likely to be violated because of intertrait dependence over time. We provide a more general framework where the dependence of the observations from the same subject over time is not assumed to be explained completely by the random effects of the model. We propose a novel, mixed model-based approach and estimate the error-covariance structure nonparametrically under a generalized linear model framework. We use penalized splines to model the general effect of time, and we consider a Dirichlet process mixture of normal prior for the random-effects distribution. We analyze blood pressure data from the Framingham Heart Study where body mass index, gender and time are treated as covariates. We compare our method with traditional methods including parametric modeling of the random effects and independent residual errors over time. We conduct extensive simulation studies to investigate the practical usefulness of the proposed method. The current approach is very helpful in analyzing bivariate irregular longitudinal traits.

Original languageEnglish (US)
Pages (from-to)3899-3910
Number of pages12
JournalStatistics in Medicine
Volume32
Issue number22
DOIs
StatePublished - Sep 30 2013

Fingerprint

Sparse Data
Semiparametric Model
Bayesian Model
Longitudinal Data
Random Effects
Mixture of Dirichlet Processes
Parametric Modeling
Penalized Splines
Mixed Effects Model
Blood Pressure
Covariance Structure
Mixed Model
Generalized Linear Model
Covariates
Irregular
Linear Models
Body Mass Index
Likely
Simulation Study
Model-based

All Science Journal Classification (ASJC) codes

  • Epidemiology
  • Statistics and Probability

Cite this

Das, Kiranmoy ; Li, Runze ; Sengupta, Subhajit ; Wu, Rongling. / A Bayesian semiparametric model for bivariate sparse longitudinal data. In: Statistics in Medicine. 2013 ; Vol. 32, No. 22. pp. 3899-3910.
@article{874b2631b0c54cd3afd22481b053af0f,
title = "A Bayesian semiparametric model for bivariate sparse longitudinal data",
abstract = "Mixed-effects models have recently become popular for analyzing sparse longitudinal data that arise naturally in biological, agricultural and biomedical studies. Traditional approaches assume independent residuals over time and explain the longitudinal dependence by random effects. However, when bivariate or multivariate traits are measured longitudinally, this fundamental assumption is likely to be violated because of intertrait dependence over time. We provide a more general framework where the dependence of the observations from the same subject over time is not assumed to be explained completely by the random effects of the model. We propose a novel, mixed model-based approach and estimate the error-covariance structure nonparametrically under a generalized linear model framework. We use penalized splines to model the general effect of time, and we consider a Dirichlet process mixture of normal prior for the random-effects distribution. We analyze blood pressure data from the Framingham Heart Study where body mass index, gender and time are treated as covariates. We compare our method with traditional methods including parametric modeling of the random effects and independent residual errors over time. We conduct extensive simulation studies to investigate the practical usefulness of the proposed method. The current approach is very helpful in analyzing bivariate irregular longitudinal traits.",
author = "Kiranmoy Das and Runze Li and Subhajit Sengupta and Rongling Wu",
year = "2013",
month = "9",
day = "30",
doi = "10.1002/sim.5790",
language = "English (US)",
volume = "32",
pages = "3899--3910",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "22",

}

A Bayesian semiparametric model for bivariate sparse longitudinal data. / Das, Kiranmoy; Li, Runze; Sengupta, Subhajit; Wu, Rongling.

In: Statistics in Medicine, Vol. 32, No. 22, 30.09.2013, p. 3899-3910.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Bayesian semiparametric model for bivariate sparse longitudinal data

AU - Das, Kiranmoy

AU - Li, Runze

AU - Sengupta, Subhajit

AU - Wu, Rongling

PY - 2013/9/30

Y1 - 2013/9/30

N2 - Mixed-effects models have recently become popular for analyzing sparse longitudinal data that arise naturally in biological, agricultural and biomedical studies. Traditional approaches assume independent residuals over time and explain the longitudinal dependence by random effects. However, when bivariate or multivariate traits are measured longitudinally, this fundamental assumption is likely to be violated because of intertrait dependence over time. We provide a more general framework where the dependence of the observations from the same subject over time is not assumed to be explained completely by the random effects of the model. We propose a novel, mixed model-based approach and estimate the error-covariance structure nonparametrically under a generalized linear model framework. We use penalized splines to model the general effect of time, and we consider a Dirichlet process mixture of normal prior for the random-effects distribution. We analyze blood pressure data from the Framingham Heart Study where body mass index, gender and time are treated as covariates. We compare our method with traditional methods including parametric modeling of the random effects and independent residual errors over time. We conduct extensive simulation studies to investigate the practical usefulness of the proposed method. The current approach is very helpful in analyzing bivariate irregular longitudinal traits.

AB - Mixed-effects models have recently become popular for analyzing sparse longitudinal data that arise naturally in biological, agricultural and biomedical studies. Traditional approaches assume independent residuals over time and explain the longitudinal dependence by random effects. However, when bivariate or multivariate traits are measured longitudinally, this fundamental assumption is likely to be violated because of intertrait dependence over time. We provide a more general framework where the dependence of the observations from the same subject over time is not assumed to be explained completely by the random effects of the model. We propose a novel, mixed model-based approach and estimate the error-covariance structure nonparametrically under a generalized linear model framework. We use penalized splines to model the general effect of time, and we consider a Dirichlet process mixture of normal prior for the random-effects distribution. We analyze blood pressure data from the Framingham Heart Study where body mass index, gender and time are treated as covariates. We compare our method with traditional methods including parametric modeling of the random effects and independent residual errors over time. We conduct extensive simulation studies to investigate the practical usefulness of the proposed method. The current approach is very helpful in analyzing bivariate irregular longitudinal traits.

UR - http://www.scopus.com/inward/record.url?scp=84884202471&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884202471&partnerID=8YFLogxK

U2 - 10.1002/sim.5790

DO - 10.1002/sim.5790

M3 - Article

C2 - 23553747

AN - SCOPUS:84884202471

VL - 32

SP - 3899

EP - 3910

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 22

ER -