Open Access
2014 A unifying approach to the estimation of the conditional Akaike information in generalized linear mixed models
Benjamin Saefken, Thomas Kneib, Clara-Sophie van Waveren, Sonja Greven
Electron. J. Statist. 8(1): 201-225 (2014). DOI: 10.1214/14-EJS881

Abstract

The conditional Akaike information criterion, AIC, has been frequently used for model selection in linear mixed models. We develop a general framework for the calculation of the conditional AIC for different exponential family distributions. This unified framework incorporates the conditional AIC for the Gaussian case, gives a new justification for Poisson distributed data and yields a new conditional AIC for exponentially distributed responses but cannot be applied to the binomial and gamma distributions. The proposed conditional Akaike information criteria are unbiased for finite samples, do not rely on a particular estimation method and do not assume that the variance-covariance matrix of the random effects is known. The theoretical results are investigated in a simulation study. The practical use of the method is illustrated by application to a data set on tree growth.

Citation

Download Citation

Benjamin Saefken. Thomas Kneib. Clara-Sophie van Waveren. Sonja Greven. "A unifying approach to the estimation of the conditional Akaike information in generalized linear mixed models." Electron. J. Statist. 8 (1) 201 - 225, 2014. https://doi.org/10.1214/14-EJS881

Information

Published: 2014
First available in Project Euclid: 27 February 2014

zbMATH: 1282.62168
MathSciNet: MR3178544
Digital Object Identifier: 10.1214/14-EJS881

Subjects:
Primary: 62J12
Secondary: 62J07

Keywords: Conditional Akaike information criterion , generalized linear mixed models , Kullback-Leibler distance , Model selection , random effects

Rights: Copyright © 2014 The Institute of Mathematical Statistics and the Bernoulli Society

Vol.8 • No. 1 • 2014
Back to Top