Some connections between Bayesian and non-Bayesian methods for regression model selection

Please use this identifier to cite or link to this item: https://doi.org/10.1016/S0167-7152(02)00048-2

Title:	Some connections between Bayesian and non-Bayesian methods for regression model selection
Authors:	Liang, F.
Keywords:	Bayes factor FPE2 criterion Kullback-Leibler distance MAP Variable selection
Issue Date:	1-Mar-2002
Citation:	Liang, F. (2002-03-01). Some connections between Bayesian and non-Bayesian methods for regression model selection. Statistics and Probability Letters 57 (1) : 53-63. ScholarBank@NUS Repository. https://doi.org/10.1016/S0167-7152(02)00048-2
Abstract:	In this article, we study the connections between Bayesian methods and non-Bayesian methods for variable selection in multiple linear regression. We show that each of the non-Bayesian criteria, FPEα, AIC, Cp and adjusted ℝ2, has its Bayesian correspondence under an appropriate prior setting. The theoretical results are illustrated by numerical simulations. © 2002 Elsevier Science B.V. All rights reserved.
Source Title:	Statistics and Probability Letters
URI:	http://scholarbank.nus.edu.sg/handle/10635/105378
ISSN:	01677152
DOI:	10.1016/S0167-7152(02)00048-2
Appears in Collections:	Staff Publications

There are no files associated with this item.

Check