Open Access
April 2014 Bayesian variable selection with shrinking and diffusing priors
Naveen Naidu Narisetty, Xuming He
Ann. Statist. 42(2): 789-817 (April 2014). DOI: 10.1214/14-AOS1207

Abstract

We consider a Bayesian approach to variable selection in the presence of high dimensional covariates based on a hierarchical model that places prior distributions on the regression coefficients as well as on the model space. We adopt the well-known spike and slab Gaussian priors with a distinct feature, that is, the prior variances depend on the sample size through which appropriate shrinkage can be achieved. We show the strong selection consistency of the proposed method in the sense that the posterior probability of the true model converges to one even when the number of covariates grows nearly exponentially with the sample size. This is arguably the strongest selection consistency result that has been available in the Bayesian variable selection literature; yet the proposed method can be carried out through posterior sampling with a simple Gibbs sampler. Furthermore, we argue that the proposed method is asymptotically similar to model selection with the $L_{0}$ penalty. We also demonstrate through empirical work the fine performance of the proposed approach relative to some state of the art alternatives.

Citation

Download Citation

Naveen Naidu Narisetty. Xuming He. "Bayesian variable selection with shrinking and diffusing priors." Ann. Statist. 42 (2) 789 - 817, April 2014. https://doi.org/10.1214/14-AOS1207

Information

Published: April 2014
First available in Project Euclid: 20 May 2014

zbMATH: 1302.62158
MathSciNet: MR3210987
Digital Object Identifier: 10.1214/14-AOS1207

Subjects:
Primary: 62F12 , 62F15 , 62J05

Keywords: Bayes factor , hierarchical model , high dimensional data , shrinkage , Variable selection

Rights: Copyright © 2014 Institute of Mathematical Statistics

Vol.42 • No. 2 • April 2014
Back to Top