Bayesian Tobit Quantile Regression Model Using Double Adaptive elastic net and Adaptive Ridge Regression

Abstract

Abstract: Recently Tobit Quantile Regression(TQR) has emerged as an important tool in statistical analysis . in order to improve the parameter estimation in (TQR) we proposed Bayesian hierarchical model with double adaptive elastic net technique and Bayesian hierarchical model with adaptive ridge regression technique . in double adaptive elastic net technique we assume different penalization parameters for penalization different regression coefficients in both parameters λ1and λ2 , also in adaptive ridge regression technique we assume different penalization parameters for penalization different regression coefficients in parameter λ . Simulation study was used for explain the efficiency of the proposed methods .The result illustrated the efficiency of the proposed methods for dealing with the estimation of parameters model in present of high correlation in explanatory variables . This is the first work that is discussing the parameter estimation in TQR model with double adaptive elastic net and adaptive ridge regression.