Bayesian Variable Selection for Zero-inflated Longitudinal Count Data

Abstract

In this paper, we consider Bayesian variable selection in the special cases of the zero-inflated power series model, viz., zero-inflated Poisson and negative binomial models for zero-inflated longitudinal count data. We propose continuous spike and Dirac spike priors to estimate the regression parameters and to select the important covariate variables simultaneously. We apply the MCMC method for posterior inference. Some simulation studies are performed to investigate the performance of the proposed approach, also, it is applied to analyze a real data set in RAND health insurance experiment data.