Model selection in binary and tobit quantile regression using the Gibbs sampler

A stochastic search variable selection approach is proposed for Bayesian model selection in binary and tobit quantile regression. A simple and efficient Gibbs sampling algorithm was developed for posterior inference using a location-scale mixture representation of the asymmetric Laplace distribution. The proposed approach is then illustrated via five simulated examples and two real data sets. Results show that the proposed method performs very well under a variety of scenarios, such as the presence of a moderately large number of covariates, collinearity and heterogeneity.