Online Program Home
My Program

Abstract Details

Activity Number: 311
Type: Contributed
Date/Time: Tuesday, August 2, 2016 : 8:30 AM to 10:20 AM
Sponsor: Biometrics Section
Abstract #318875
Title: Comparative GQL and GMM Bootstrap Methods for Hierarchical Data
Author(s): Bei Wang*
Companies: Arizona State University
Keywords: Generalized linear mixed model ; Correlated data ; GMM ; GQL ; Resample schemes

Though the likelihood is a useful tool for obtaining estimation of the distribution underlying the sample and the parameters of interest that are derived from this distribution it is not readily available in the fit of generalized linear mixed models. Various alternative methods such as generalized quasi-likelihood (GQL) and generalized method of moments (GMM) which do not require complete knowledge of the distribution of the data have been proposed. In addition, bootstrapping procedure can provide distribution-independent method to further assess the properties of estimators. In this paper, we present a GQL bootstrap logistic regression model and a GMM bootstrap logistic regression models with representative resampling strategies as a means of analyzing binary data obtained from a hierarchical structure. We examine and compare the performance of both models combined with several resampling schemes and the fit of these scenarios with multiple random effects. A simulation study is conducted and a numerical example with three levels of nesting is analyzed.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association