Simple marginally noninformative prior distributions for covariance matrices

Huang, Alan and Wand, M. P. (2013) Simple marginally noninformative prior distributions for covariance matrices. Bayesian Analysis, 8 2: 439-452. doi:10.1214/13-BA815

Author Huang, Alan
Wand, M. P.
Title Simple marginally noninformative prior distributions for covariance matrices
Journal name Bayesian Analysis   Check publisher's open access policy
ISSN 1936-0975
Publication date 2013-01-01
Sub-type Article (original research)
DOI 10.1214/13-BA815
Open Access Status DOI
Volume 8
Issue 2
Start page 439
End page 452
Total pages 14
Place of publication Pittsburgh, PA, United States
Publisher International Society for Bayesian Analysis
Language eng
Abstract A family of prior distributions for covariance matrices is studied. Members of the family possess the attractive property of all standard deviation and correlation parameters being marginally noninformative for particular hyper-parameter choices. Moreover, the family is quite simple and, for approximate Bayesian inference techniques such as Markov chain Monte Carlo and mean eld variational Bayes, has tractability on par with the Inverse-Wishart conjugate fam-ily of prior distributions. A simulation study shows that the new prior distributions can lead to more accurate sparse covariance matrix estimation.
Keyword Bayesian inference
Gibbs sampling
Markov chain monte carlo
Mean field variational bayes
Q-Index Code C1
Q-Index Status Confirmed Code
Institutional Status Non-UQ

Document type: Journal Article
Sub-type: Article (original research)
Collections: School of Mathematics and Physics
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 27 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 30 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Thu, 27 Mar 2014, 00:09:12 EST by Kay Mackie on behalf of Scholarly Communication and Digitisation Service