Title: Bayesian variable selection using Lasso
Speaker: Yuchen Han (Case Western Reserve University)
Advisor: Anirban Mondal (Assistant Professor, Case Western Reserve University)
Abstract: This thesis proposes to combine the stochastic search variable selection (SSVS) approach and Bayesian Lasso approach by introducing a double exponential distribution on the conditional prior of the regression parameters given the indication variables. Gibbs Sampling would be used to sample from the joint posterior distribution. We compare this new method to existing Bayesian variable selection methods such as Kou and Mallick and George and McCullochand provide an overall qualitative assessment of the computational speed and efficiency of mixing and separation.