BAYESIAN RECIPROCAL ADAPTIVE LASSO QUANTILE REGRESSION: SIMULATION AND REAL DATA ANALYSIS

Abstract

Regression analyses concerns in the explanation and the prediction of the relation between the response variable and a set of predictor variables. The explanation of the regression model obtained via variable selection procedure, while the prediction accuracy of the regression model of the obtained by trading off the bias and the variance of the estimator. This paper discuss the employing the Bayesian adaptive lasso penalized function in quantile regression. The reciprocal adaptive lasso works as variable selection procedure. We employed the scale mixture of normals and the scale mixture of uniforms to develop the Gibbs sampler algorithm. Full conditional posterior distributions have derived based of the hierarchical prior model. Two simulation scenarios and real data analysis conducted to test the performance of the two proposed Bayesian adaptive lasso methods in quantile regression.