Department of Statistics, School of Management, Fudan University

In this paper, we compare two modified
Gaussian pseudolikelihood criteria (GPCs) with existing Gaussian
pseudolikelihood criterion and empirical likelihood based criteria
to choose the working correlation matrix in generalized estimating
equations approach. Rich simulation studies are conducted to
investigate the performance of these criteria under a range of model
settings. The results show that the modified criteria outperform the
original GPC and empirical likelihood based criteria in most cases
in terms of selection accuracy. Empirical likelihood based criteria
perform better to identify exchangeable structure in data with
binary response. In the end, these criteria are applied to epilepsy
seizure and Madras longitudinal schizophrenia study clinical data
sets analysis.