Bayesian covariance matrix estimation using a mixture of decomposable graphical models
- First Online:
We present a Bayesian approach to estimating a covariance matrix by using a prior that is a mixture over all decomposable graphs, with the probability of each graph size specified by the user and graphs of equal size assigned equal probability. Most previous approaches assume that all graphs are equally probable. We show empirically that the prior that assigns equal probability over graph sizes outperforms the prior that assigns equal probability over all graphs in more efficiently estimating the covariance matrix. The prior requires knowing the number of decomposable graphs for each graph size and we give a simulation method for estimating these counts. We also present a Markov chain Monte Carlo method for estimating the posterior distribution of the covariance matrix that is much more efficient than current methods. Both the prior and the simulation method to evaluate the prior apply generally to any decomposable graphical model.
KeywordsCovariance selection Reduced conditional sampling Variable selection
Unable to display preview. Download preview PDF.
- Giudici, P.: Learning in graphical Gaussian models. In: Berger, A.P.D.J., Bernardo, J.M., Smith, A.F.M. (eds.) Bayesian Statistics 5: Proceedings of the Fifth Valencia International Meeting, June 5–9, 1994, vol. 5, pp. 621–628. Oxford University Press, London (1996) Google Scholar
- Larner, M.: Mass and its relationship to physical measurements. Technical Report, Department of Mathematics, University of Queensland, Australia (1996) Google Scholar
- Lauritzen, S.L.: Graphical Models. Oxford University Press, London (1996) Google Scholar
- Whittaker, J.: Graphical Models in Applied Mathematical Analysis. Wiley, New York (1990) Google Scholar