Bayesian adaptive lasso for the generalized Poisson hurdle model

Variable selection is important in statistical inference. In health economics and management, count outcomes are frequently encountered and often have a large proportion of zeros, and these data change the relation between mean and variance in the Poisson distribution setting. In this paper, for the generalized Poisson hurdle model, which can flexibly fit zero-inflated data with dispersion characteristics, we propose a Bayesian adaptive lasso method to conduct a simultaneous variable selection and parameter estimation. Efficient MCMC methods are applied to carry out posterior sampling and inference. Moreover, we investigate the finite sample performance of the proposed method through a simulation study and apply it to analyze real-life data about doctor visits.