an:05915060
Zbl 1216.62039
Freeman, G.; Smith, J. Q.
Bayesian MAP model selection of chain event graphs
EN
J. Multivariate Anal. 102, No. 7, 1152-1165 (2011).
00281674
2011
j
62F15 68T35 05C90 62-07
Bayesian model selection; Dirichlet distribution
Summary: Chain event graphs are graphical models that while retaining most of the structural advantages of Bayesian networks for model interrogation, propagation and learning, more naturally encode asymmetric state spaces and the order in which events happen than Bayesian networks do. In addition, the class of models that can be represented by chain event graphs for a finite set of discrete variables is a strict superset of the class that can be described by Bayesian networks. In this paper we demonstrate how with complete sampling, conjugate closed form model selection based on product Dirichlet priors is possible, and prove that suitable homogeneity assumptions characterise the product Dirichlet prior on this class of models. We demonstrate our techniques using two educational examples.