Model Selection for Zero-Inflated Generalized Linear Models

• Activity Number: 321 - Machine Learning and Variable Selection
• Type: Contributed
• Date/Time: Wednesday, August 11, 2021 : 3:30 PM to 5:20 PM
• Sponsor: Section on Statistical Computing
• Author(s): Abdulla Al Mamun* and Sudhir Paul
• Companies: Gonzaga University and University of Windsor
• Keywords: Generalized linear model; Zero-inflation; Over-dispersion; Score test; Forward Selection method

Abstract

Model selection is a necessary step in many practical regression analysis that provides a good, yet parsimonious, model for the response. In this paper model selection procedure is developed in zero-inflated generalized linear models (ZIGLM). The main tool that is being used is the score test, other large sample tests, such as, the likelihood ratio (LR) and the Wald test, the AIC and the BIC are included in the comparison. A simulation study is conducted to examine the performance of the test procedure in terms of empirical level and power. For illustration, the method is applied to one health data set.