A likelihood reformulation method in non-normal random effects models

Lei Liu, Zhangsheng Yu

Research output: Contribution to journalArticle

46 Citations (Scopus)

Abstract

In this paper, we propose a practical computational method to obtain the maximum likelihood estimates (MLE) for mixed models with non-normal random effects. By simply multiplying and dividing a standard normal density, we reformulate the likelihood conditional on the non-normal random effects to that conditional on the normal random effects. Gaussian quadrature technique, conveniently implemented in SAS Proc NLMIXED, can then be used to carry out the estimation process. Our method substantially reduces computational time, while yielding similar estimates to the probability integral transformation method (J. Comput. Graphical Stat. 2006; 15:39-57). Furthermore, our method can be applied to more general situations, e.g. finite mixture random effects or correlated random effects from Clayton copula. Simulations and applications are presented to illustrate our method.

Original languageEnglish (US)
Pages (from-to)3105-3124
Number of pages20
JournalStatistics in Medicine
Volume27
Issue number16
DOIs
StatePublished - Jul 20 2008
Externally publishedYes

Fingerprint

Random Effects Model
Random Effects
Reformulation
Likelihood
Likelihood Functions
Conditional Likelihood
Integral Transformation
Gaussian Quadrature
Finite Mixture
Mixed Model
Copula
Maximum Likelihood Estimate
Computational Methods
Estimate
Simulation

Keywords

  • Gamma frailty model
  • Gaussian copula
  • Generalized linear mixed model
  • Heterogeneity model
  • Logistic distribution

ASJC Scopus subject areas

  • Epidemiology

Cite this

A likelihood reformulation method in non-normal random effects models. / Liu, Lei; Yu, Zhangsheng.

In: Statistics in Medicine, Vol. 27, No. 16, 20.07.2008, p. 3105-3124.

Research output: Contribution to journalArticle

@article{51ebd43809f0464280087268e282aae4,
title = "A likelihood reformulation method in non-normal random effects models",
abstract = "In this paper, we propose a practical computational method to obtain the maximum likelihood estimates (MLE) for mixed models with non-normal random effects. By simply multiplying and dividing a standard normal density, we reformulate the likelihood conditional on the non-normal random effects to that conditional on the normal random effects. Gaussian quadrature technique, conveniently implemented in SAS Proc NLMIXED, can then be used to carry out the estimation process. Our method substantially reduces computational time, while yielding similar estimates to the probability integral transformation method (J. Comput. Graphical Stat. 2006; 15:39-57). Furthermore, our method can be applied to more general situations, e.g. finite mixture random effects or correlated random effects from Clayton copula. Simulations and applications are presented to illustrate our method.",
keywords = "Gamma frailty model, Gaussian copula, Generalized linear mixed model, Heterogeneity model, Logistic distribution",
author = "Lei Liu and Zhangsheng Yu",
year = "2008",
month = "7",
day = "20",
doi = "10.1002/sim.3153",
language = "English (US)",
volume = "27",
pages = "3105--3124",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "16",

}

TY - JOUR

T1 - A likelihood reformulation method in non-normal random effects models

AU - Liu, Lei

AU - Yu, Zhangsheng

PY - 2008/7/20

Y1 - 2008/7/20

N2 - In this paper, we propose a practical computational method to obtain the maximum likelihood estimates (MLE) for mixed models with non-normal random effects. By simply multiplying and dividing a standard normal density, we reformulate the likelihood conditional on the non-normal random effects to that conditional on the normal random effects. Gaussian quadrature technique, conveniently implemented in SAS Proc NLMIXED, can then be used to carry out the estimation process. Our method substantially reduces computational time, while yielding similar estimates to the probability integral transformation method (J. Comput. Graphical Stat. 2006; 15:39-57). Furthermore, our method can be applied to more general situations, e.g. finite mixture random effects or correlated random effects from Clayton copula. Simulations and applications are presented to illustrate our method.

AB - In this paper, we propose a practical computational method to obtain the maximum likelihood estimates (MLE) for mixed models with non-normal random effects. By simply multiplying and dividing a standard normal density, we reformulate the likelihood conditional on the non-normal random effects to that conditional on the normal random effects. Gaussian quadrature technique, conveniently implemented in SAS Proc NLMIXED, can then be used to carry out the estimation process. Our method substantially reduces computational time, while yielding similar estimates to the probability integral transformation method (J. Comput. Graphical Stat. 2006; 15:39-57). Furthermore, our method can be applied to more general situations, e.g. finite mixture random effects or correlated random effects from Clayton copula. Simulations and applications are presented to illustrate our method.

KW - Gamma frailty model

KW - Gaussian copula

KW - Generalized linear mixed model

KW - Heterogeneity model

KW - Logistic distribution

UR - http://www.scopus.com/inward/record.url?scp=48049093037&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=48049093037&partnerID=8YFLogxK

U2 - 10.1002/sim.3153

DO - 10.1002/sim.3153

M3 - Article

C2 - 18038445

AN - SCOPUS:48049093037

VL - 27

SP - 3105

EP - 3124

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 16

ER -