Abstract
bstract : Quantile regression extends the mean regression model to conditional quantiles of the response variable (Koenker and Bassett, 1978; Koenker, 2005). As a result of its capacity to take advantage of all available data in the analysis, the Bayesian method of analysis has become extremely popular. We proposed a novel estimation method and selected variables. Using Bayesian methodology for binary quantile regression. On the basis of the recently proposed scale mixture of normal distribution mixing Rayleigh density representation for Laplace distribution prior density of the parameters vector, new hierarchical models have been developed. Simulation Scenarios and actual data are used to evaluate the efficacy of the new method in comparison to two existing methods.
