|
Activity Number:
|
1
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, July 29, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #308090 |
|
Title:
|
Flexible Priors for Covariance Matrices in Gaussian Graphical Models
|
|
Author(s):
|
Bala Rajaratnam*+ and Carlos Carvalho and Helene M. Massam
|
|
Companies:
|
Statistical and Applied Mathematical Sciences Institute and Duke University and York University
|
|
Address:
|
19 T.W. Alexander Dr., Durham, NC, 27709,
|
|
Keywords:
|
|
|
Abstract:
|
The hyper-inverse Wishart distribution is a commonly used prior for Bayesian inference on covariance matrices in Gaussian Graphical models. This prior has the distinct advantage that it is the conjugate prior for this model but suffers from lack of flexibility in high dimensional problems due to its single shape parameter. In this paper, we propose flexible classes of priors for posterior inference on covariance matrices in Gaussian Graphical models that allow for up to k+1 shape parameters where k denotes the number of cliques in the graph. We investigate the corresponding Bayes estimators under usual losses considered in the literature and exploit the conjugacy relationship in these models to express Bayes estimators in closed form. The closed form solutions allow us to avoid heavy computational costs that are usually incurred in these problems.
|
