Bayesian group Lasso regression for left-censored data

Abstract

In this paper, a new approach for model selection in left-censored regression has been presented. Specifically, we proposed a new Bayesian group Lasso for variable selection and coefficient estimation in left-censored data (BGLRLC). A new hierarchical Bayesian modeling for group Lasso has introduced, which motivate us to propose a new Gibbs sampler for sampling the parameters from the posteriors. The performance of the proposed approach is examined through simulation studies and a real data analysis. Results show that the proposed approach performs well in comparison to other existing methods.