Structuring Shrinkage: Some Correlated Priors for Regression

Griffin, Jim E. and Brown, Philip J. (2012) Structuring Shrinkage: Some Correlated Priors for Regression. Biometrika, 99 (2). pp. 481-487. ISSN 1464-3510. (The full text of this publication is not available from this repository)

The full text of this publication is not available from this repository. (Contact us about this Publication)
Official URL
http://dx.doi.org/10.1093/biomet/asr082

Abstract

This paper develops a rich class of sparsity priors for regression effects that encourage shrinkage of both regression effects and contrasts between effects to zerowhilst leaving sizeable real effects largely unshrunk. The construction of these priors uses some properties of normal-gamma distributions to include design features in the prior specification, but has general relevance to any continuous sparsity prior. Specific prior distributions are developed for serial dependence between regression effects and correlation within groups of regression effects.

Item Type: Article
Uncontrolled keywords: Fused prior, Grouped prior, Lasso, Multiple regression, Normal-gamma prior, Sparsity
Subjects: Q Science > QA Mathematics (inc Computing science) > QA276 Mathematical statistics
Divisions: Faculties > Science Technology and Medical Studies > School of Mathematics Statistics and Actuarial Science > Statistics
Depositing User: Jim Griffin
Date Deposited: 30 May 2012 12:45
Last Modified: 13 May 2014 11:09
Resource URI: http://kar.kent.ac.uk/id/eprint/29600 (The current URI for this page, for reference purposes)
  • Depositors only (login required):