A computationally tractable multivariate random effects model for clustered binary data

Brent A. Coull, E. Andres Houseman, Rebecca Betensky

Research output: Contribution to journalArticle

Abstract

We consider a multivariate random effects model for clustered binary data that is useful when interest focuses on the association structure among clustered observations. Based on a vector of gamma random effects and a complementary log-log link function, the model yields a likelihood that has closed form, making a frequentist approach to model-fitting straightforward. This closed form yields several advantages over existing methods, including easy inspection of model identifiability and straightforward adjustment for nonrandom ascertainment of subjects, such as that which occurs in family studies of disease aggregation. We use the proposed model to analyse two different binary datasets concerning disease outcome data from a familial aggregation study of breast and ovarian cancer in women and loss of heterozygosity outcomes from a brain tumour study.

Original languageEnglish (US)
Pages (from-to)587-599
Number of pages13
JournalBiometrika
Volume93
Issue number3
DOIs
StatePublished - Sep 1 2006

Fingerprint

Clustered Data
Random Effects Model
Binary Data
Multivariate Models
Loss of Heterozygosity
Aggregation
Closed-form
Brain Neoplasms
Ovarian Neoplasms
Ovarian Cancer
Brain Tumor
Link Function
Model Fitting
Identifiability
Breast Neoplasms
Random Effects
Breast Cancer
Agglomeration
Inspection
Likelihood

Keywords

  • Binary time series
  • Complementary log-log link
  • Generalised linear mixed model
  • Multivariate gamma

ASJC Scopus subject areas

  • Statistics and Probability
  • Mathematics(all)
  • Agricultural and Biological Sciences (miscellaneous)
  • Agricultural and Biological Sciences(all)
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

A computationally tractable multivariate random effects model for clustered binary data. / Coull, Brent A.; Houseman, E. Andres; Betensky, Rebecca.

In: Biometrika, Vol. 93, No. 3, 01.09.2006, p. 587-599.

Research output: Contribution to journalArticle

Coull, Brent A. ; Houseman, E. Andres ; Betensky, Rebecca. / A computationally tractable multivariate random effects model for clustered binary data. In: Biometrika. 2006 ; Vol. 93, No. 3. pp. 587-599.
@article{9ce8d932aef9499f8d3242e50c0cc0b2,
title = "A computationally tractable multivariate random effects model for clustered binary data",
abstract = "We consider a multivariate random effects model for clustered binary data that is useful when interest focuses on the association structure among clustered observations. Based on a vector of gamma random effects and a complementary log-log link function, the model yields a likelihood that has closed form, making a frequentist approach to model-fitting straightforward. This closed form yields several advantages over existing methods, including easy inspection of model identifiability and straightforward adjustment for nonrandom ascertainment of subjects, such as that which occurs in family studies of disease aggregation. We use the proposed model to analyse two different binary datasets concerning disease outcome data from a familial aggregation study of breast and ovarian cancer in women and loss of heterozygosity outcomes from a brain tumour study.",
keywords = "Binary time series, Complementary log-log link, Generalised linear mixed model, Multivariate gamma",
author = "Coull, {Brent A.} and Houseman, {E. Andres} and Rebecca Betensky",
year = "2006",
month = "9",
day = "1",
doi = "10.1093/biomet/93.3.587",
language = "English (US)",
volume = "93",
pages = "587--599",
journal = "Biometrika",
issn = "0006-3444",
publisher = "Oxford University Press",
number = "3",

}

TY - JOUR

T1 - A computationally tractable multivariate random effects model for clustered binary data

AU - Coull, Brent A.

AU - Houseman, E. Andres

AU - Betensky, Rebecca

PY - 2006/9/1

Y1 - 2006/9/1

N2 - We consider a multivariate random effects model for clustered binary data that is useful when interest focuses on the association structure among clustered observations. Based on a vector of gamma random effects and a complementary log-log link function, the model yields a likelihood that has closed form, making a frequentist approach to model-fitting straightforward. This closed form yields several advantages over existing methods, including easy inspection of model identifiability and straightforward adjustment for nonrandom ascertainment of subjects, such as that which occurs in family studies of disease aggregation. We use the proposed model to analyse two different binary datasets concerning disease outcome data from a familial aggregation study of breast and ovarian cancer in women and loss of heterozygosity outcomes from a brain tumour study.

AB - We consider a multivariate random effects model for clustered binary data that is useful when interest focuses on the association structure among clustered observations. Based on a vector of gamma random effects and a complementary log-log link function, the model yields a likelihood that has closed form, making a frequentist approach to model-fitting straightforward. This closed form yields several advantages over existing methods, including easy inspection of model identifiability and straightforward adjustment for nonrandom ascertainment of subjects, such as that which occurs in family studies of disease aggregation. We use the proposed model to analyse two different binary datasets concerning disease outcome data from a familial aggregation study of breast and ovarian cancer in women and loss of heterozygosity outcomes from a brain tumour study.

KW - Binary time series

KW - Complementary log-log link

KW - Generalised linear mixed model

KW - Multivariate gamma

UR - http://www.scopus.com/inward/record.url?scp=33750156714&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33750156714&partnerID=8YFLogxK

U2 - 10.1093/biomet/93.3.587

DO - 10.1093/biomet/93.3.587

M3 - Article

AN - SCOPUS:33750156714

VL - 93

SP - 587

EP - 599

JO - Biometrika

JF - Biometrika

SN - 0006-3444

IS - 3

ER -