Simultaneous variable selection for joint models of longitudinal and survival outcomes

Zangdong He, Wanzhu Tu, Sijian Wang, Haoda Fu, Zhangsheng Yu

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Joint models of longitudinal and survival outcomes have been used with increasing frequency in clinical investigations. Correct specification of fixed and random effects is essential for practical data analysis. Simultaneous selection of variables in both longitudinal and survival components functions as a necessary safeguard against model misspecification. However, variable selection in such models has not been studied. No existing computational tools, to the best of our knowledge, have been made available to practitioners. In this article, we describe a penalized likelihood method with adaptive least absolute shrinkage and selection operator (ALASSO) penalty functions for simultaneous selection of fixed and random effects in joint models. To perform selection in variance components of random effects, we reparameterize the variance components using a Cholesky decomposition; in doing so, a penalty function of group shrinkage is introduced. To reduce the estimation bias resulted from penalization, we propose a two-stage selection procedure in which the magnitude of the bias is ameliorated in the second stage. The penalized likelihood is approximated by Gaussian quadrature and optimized by an EM algorithm. Simulation study showed excellent selection results in the first stage and small estimation biases in the second stage. To illustrate, we analyzed a longitudinally observed clinical marker and patient survival in a cohort of patients with heart failure.

Original languageEnglish
Pages (from-to)178-187
Number of pages10
JournalBiometrics
Volume71
Issue number1
DOIs
StatePublished - Mar 1 2015

Fingerprint

Joint Model
Variable Selection
Joints
Random Effects
Penalized Likelihood
Survival
Fixed Effects
Penalty Function
Shrinkage
shrinkage
Cholesky Decomposition
Heart Failure
Selection of Variables
Two-stage Procedure
Components of Variance
Gaussian Quadrature
Model Misspecification
Variance Components
Penalization
Likelihood Methods

Keywords

  • ALASSO
  • Cholesky decomposition
  • EM algorithm
  • Gaussian quadrature
  • Joint models
  • Mixed effect selection

ASJC Scopus subject areas

  • Applied Mathematics
  • Statistics and Probability
  • Agricultural and Biological Sciences(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Immunology and Microbiology(all)
  • Medicine(all)

Cite this

Simultaneous variable selection for joint models of longitudinal and survival outcomes. / He, Zangdong; Tu, Wanzhu; Wang, Sijian; Fu, Haoda; Yu, Zhangsheng.

In: Biometrics, Vol. 71, No. 1, 01.03.2015, p. 178-187.

Research output: Contribution to journalArticle

He, Zangdong ; Tu, Wanzhu ; Wang, Sijian ; Fu, Haoda ; Yu, Zhangsheng. / Simultaneous variable selection for joint models of longitudinal and survival outcomes. In: Biometrics. 2015 ; Vol. 71, No. 1. pp. 178-187.
@article{2614e5301ff74b068f60311c4355bcb4,
title = "Simultaneous variable selection for joint models of longitudinal and survival outcomes",
abstract = "Joint models of longitudinal and survival outcomes have been used with increasing frequency in clinical investigations. Correct specification of fixed and random effects is essential for practical data analysis. Simultaneous selection of variables in both longitudinal and survival components functions as a necessary safeguard against model misspecification. However, variable selection in such models has not been studied. No existing computational tools, to the best of our knowledge, have been made available to practitioners. In this article, we describe a penalized likelihood method with adaptive least absolute shrinkage and selection operator (ALASSO) penalty functions for simultaneous selection of fixed and random effects in joint models. To perform selection in variance components of random effects, we reparameterize the variance components using a Cholesky decomposition; in doing so, a penalty function of group shrinkage is introduced. To reduce the estimation bias resulted from penalization, we propose a two-stage selection procedure in which the magnitude of the bias is ameliorated in the second stage. The penalized likelihood is approximated by Gaussian quadrature and optimized by an EM algorithm. Simulation study showed excellent selection results in the first stage and small estimation biases in the second stage. To illustrate, we analyzed a longitudinally observed clinical marker and patient survival in a cohort of patients with heart failure.",
keywords = "ALASSO, Cholesky decomposition, EM algorithm, Gaussian quadrature, Joint models, Mixed effect selection",
author = "Zangdong He and Wanzhu Tu and Sijian Wang and Haoda Fu and Zhangsheng Yu",
year = "2015",
month = "3",
day = "1",
doi = "10.1111/biom.12221",
language = "English",
volume = "71",
pages = "178--187",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - Simultaneous variable selection for joint models of longitudinal and survival outcomes

AU - He, Zangdong

AU - Tu, Wanzhu

AU - Wang, Sijian

AU - Fu, Haoda

AU - Yu, Zhangsheng

PY - 2015/3/1

Y1 - 2015/3/1

N2 - Joint models of longitudinal and survival outcomes have been used with increasing frequency in clinical investigations. Correct specification of fixed and random effects is essential for practical data analysis. Simultaneous selection of variables in both longitudinal and survival components functions as a necessary safeguard against model misspecification. However, variable selection in such models has not been studied. No existing computational tools, to the best of our knowledge, have been made available to practitioners. In this article, we describe a penalized likelihood method with adaptive least absolute shrinkage and selection operator (ALASSO) penalty functions for simultaneous selection of fixed and random effects in joint models. To perform selection in variance components of random effects, we reparameterize the variance components using a Cholesky decomposition; in doing so, a penalty function of group shrinkage is introduced. To reduce the estimation bias resulted from penalization, we propose a two-stage selection procedure in which the magnitude of the bias is ameliorated in the second stage. The penalized likelihood is approximated by Gaussian quadrature and optimized by an EM algorithm. Simulation study showed excellent selection results in the first stage and small estimation biases in the second stage. To illustrate, we analyzed a longitudinally observed clinical marker and patient survival in a cohort of patients with heart failure.

AB - Joint models of longitudinal and survival outcomes have been used with increasing frequency in clinical investigations. Correct specification of fixed and random effects is essential for practical data analysis. Simultaneous selection of variables in both longitudinal and survival components functions as a necessary safeguard against model misspecification. However, variable selection in such models has not been studied. No existing computational tools, to the best of our knowledge, have been made available to practitioners. In this article, we describe a penalized likelihood method with adaptive least absolute shrinkage and selection operator (ALASSO) penalty functions for simultaneous selection of fixed and random effects in joint models. To perform selection in variance components of random effects, we reparameterize the variance components using a Cholesky decomposition; in doing so, a penalty function of group shrinkage is introduced. To reduce the estimation bias resulted from penalization, we propose a two-stage selection procedure in which the magnitude of the bias is ameliorated in the second stage. The penalized likelihood is approximated by Gaussian quadrature and optimized by an EM algorithm. Simulation study showed excellent selection results in the first stage and small estimation biases in the second stage. To illustrate, we analyzed a longitudinally observed clinical marker and patient survival in a cohort of patients with heart failure.

KW - ALASSO

KW - Cholesky decomposition

KW - EM algorithm

KW - Gaussian quadrature

KW - Joint models

KW - Mixed effect selection

UR - http://www.scopus.com/inward/record.url?scp=84926199533&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84926199533&partnerID=8YFLogxK

U2 - 10.1111/biom.12221

DO - 10.1111/biom.12221

M3 - Article

VL - 71

SP - 178

EP - 187

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 1

ER -