Objective Bayesian model selection in Gaussian graphical models
- 4 May 2009
- journal article
- research article
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 96 (3) , 497-512
Abstract
This paper presents a default model-selection procedure for Gaussian graphical models that involves two new developments. First, we develop a default version of the hyper-inverse Wishart prior for restricted covariance matrices, called the hyper-inverse Wishart g-prior, and show how it corresponds to the implied fractional prior for selecting a graph using fractional Bayes factors. Second, we apply a class of priors that automatically handles the problem of multiple hypothesis testing. We demonstrate our methods on a variety of simulated examples, concluding with a real example analyzing covariation in mutual-fund returns. These studies reveal that the combined use of a multiplicity-correction prior on graphs and fractional Bayes factors for computing marginal likelihoods yields better performance than existing Bayesian methods.Keywords
This publication has 19 references indexed in Scilit:
- Simulation of hyper-inverse Wishart distributions in graphical modelsBiometrika, 2007
- Wishart distributions for decomposable graphsThe Annals of Statistics, 2007
- Dynamic matrix-variate graphical modelsBayesian Analysis, 2007
- Experiments in Stochastic Computation for High-Dimensional Graphical ModelsStatistical Science, 2005
- A Monte Carlo method for computing the marginal likelihood in nondecomposable Gaussian graphical modelsBiometrika, 2005
- Sparse graphical models for exploring gene expression dataJournal of Multivariate Analysis, 2004
- Calibration and empirical Bayes variable selectionBiometrika, 2000
- Decomposable graphical Gaussian model determinationBiometrika, 1999
- The Intrinsic Bayes Factor for Model Selection and PredictionJournal of the American Statistical Association, 1996
- Hyper Markov Laws in the Statistical Analysis of Decomposable Graphical ModelsThe Annals of Statistics, 1993