Biometrika Advance Access originally published online on August 5, 2007
Biometrika 2007 94(3):647-659; doi:10.1093/biomet/asm056
Copyright © 2007 Biometrika Trust
Articles |
Simulation of hyper-inverse Wishart distributions in graphical models
Department of Statistical Science, Duke University, Durham, North Carolina 27708-0251, U.S.A.
Department of Mathematics & Statistics, York University, Toronto M3 J1P3, Canada
Department of Statistical Science, Duke University, Durham, North Carolina 27708-0251, U.S.A.
carlos{at}stat.duke.edu
massamh{at}mathstat.yorku.ca
mw{at}stat.duke.edu
Received for publication 1 May 2006. Revision received 1 January 2007.
We introduce and exemplify an efficient method for direct sampling from hyper-inverse Wishart distributions. The method relies very naturally on the use of standard junction-tree representation of graphs, and couples these with matrix results for inverse Wishart distributions. We describe the theory and resulting computational algorithms for both decomposable and nondecomposable graphical models. An example drawn from financial time series demonstrates application in a context where inferences on a structured covariance model are required. We discuss and investigate questions of scalability of the simulation methods to higher-dimensional distributions. The paper concludes with general comments about the approach, including its use in connection with existing Markov chain Monte Carlo methods that deal with uncertainty about the graphical model structure.
Key Words: Gaussian graphical model • Hyper-inverse Wishart • Junction tree • Portfolio analysis • Posterior simulation
References
-
Aguilar O., West M. Bayesian dynamic factor models and portfolio allocation. J. Bus. Econ. Statist. (2000) 18:338–57.[CrossRef]
Atay-Kayis A., Massam H. The marginal likelihood for decomposable and nondecomposable graphical Gaussian models. Biometrika (2005) 92:317–35.
Carvalho C. M., West M. Dynamic matrix-variate graphical models. Bayesian Anal. (2007a) 2:69–98.
Carvalho C. M., West M. Dynamic matrix-variate graphical models–a synopsis. In: Bayesian Statistics VIII—Bernardo J., Bayarri M., Berger J., Dawid A., Heckerman D., Smith A., West M., eds. (2007b) Oxford: Oxford University Press. 585–90.
Dawid A. P., Lauritzen S. L. Hyper-Markov laws in the statistical analysis of decomposable graphical models. Ann. Statist. (1993) 21:1272–317.
Dempster A. Covariance selection. Biometrics (1972) 28:157–75.[CrossRef][ISI]
Dobra A., Jones B., Hans C., Nevins J., West M.A. Sparse graphical models for exploring gene expression data. J. Mult. Anal. (2004) 90:196–212.[CrossRef]
Giudici P., Green P. J. Decomposable graphical Gaussian model determination. Biometrika (1999) 86:785–801.
Jones B., Carvalho C., Dobra A., Hans C., Carter C., West M. Experiments in stochastic computation for high-dimensional graphical models. Statist. Sci. (2005) 20:388–400.[CrossRef]
Lauritzen S. L. Graphical Models (1996) Oxford: Clarendon Press.
Ledoit O., Wolf M. Honey, I shrunk the sample covariance matrix. J. Portfolio Manag. (2004) 30:110–9.
Massam H., Neher E. Estimation and testing for lattice conditional independence models on Euclidean Jordan algebras. Ann. Statist. (1998) 26:1051–82.[CrossRef]
Quintana J., Lourdes V., Aguilar O., Liu J. Global gambling. In: Bayesian Statistics VII—Bernardo J., Bayarri M., Berger J., Dawid A., Heckerman D., Smith A., West M., eds. (2003) Oxford: Oxford University Press. 349–68.
Roverato A. Cholesky decomposition of a hyper-inverse Wishart matrix. Biometrika (2000) 87:99–112.
Simpson A., Wilkinson D. Computationally intensive techniques for a fully Bayesian, decision theoretic approach to financial forecasting and portfolio selection. Comp. Sci. Statist. (2002) 33:472–83.
Wilkinson D., Yeung S. Conditional simulation from highly structured Gaussian systems, with application to blocking-MCMC for the Bayesian analysis of very large linear models. Statist. Comp. (2002) 12:287–300.[CrossRef]
Wilkinson D., Yeung S. A sparse matrix approach to Bayesian computation in large linear models. Comp. Statist. Data Anal. (2004) 44:493–516.[CrossRef]
Wong F., Carter C., Kohn R. Efficient estimation of covariance selection models. Biometrika (2003) 90:809–30.
Yeung S., Wilkinson D. Adaptive metropolis-hastings samplers for the Bayesian analysis of large linear Gaussian systems. Comp. Sci. Statist. (2002) 33:128–38.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||