Bayesian analysis can reach the aim of model selection based on shrinkage priors. Here, we explain three methods that have been demonstrated giving reasonable performance when working on high-dimensional model selection problems. Also, we compare these methods using simulation and give real data application. In addition, we extend the method based on Dirichlet–Laplace (DL) prior from normal means problem to linear regression model, and show the minimax contraction rate still holds under mild conditions.
TibshiraniR. Regression Shrinkage and Selection via the Lasso. J Roy Stat Soc B. 1996; 267–88.
2.
LiangFTruongYKWongWH.Automatic Bayesian Model Averaging for Linear Regression and Applications in Bayesian Curve Fitting. Stat Sin. 2001; 11 (4): 1005–30.
3.
NarisettyNNHeX. Bayesian Variable Selection with Shrinking and Diffusing Priors. Ann Stat. 2014; 42 (2): 789–817.
4.
BhattacharyaAPatiDPillaiNSDunsonDB.Dirichlet-Laplace Priors for Optimal Shrinkage. J Amer Stat Assoc.2015; 110 (512): 1479–1490.
5.
JohnsonVE.On Numerical Aspects of Bayesian Model Selection in High and Ultrahigh-dimensional Settings. Baye Anal. 2013; 8 (4): 741–58.
