On the toric algebra of graphical models

Abstract

We formulate necessary and sufficient conditions for an arbitrary discrete probability distribution to factor according to an undirected graphical model, or a log-linear model, or other more general exponential models. For decomposable graphical models these conditions are equivalent to a set of conditional independence statements similar to the Hammersley--Clifford theorem; however, we show that for nondecomposable graphical models they are not. We also show that nondecomposable models can have nonrational maximum likelihood estimates. These results are used to give several novel characterizations of decomposable graphical models.

Published at http://dx.doi.org/10.1214/009053606000000263 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Cited by in corpus (16)

• Conjunctive Bayesian networks
• Markov bases for two-way subtable sum problems
• A Markov Basis for Conditional Test of Common Diagonal Effect in Quasi-Independence Model for Square Contingency Tables
• Boundary cliques, clique trees and perfect sequences of maximal cliques of a chordal graph
• Three Counterexamples on Semigraphoids
• Algebraic Bayesian analysis of contingency tables with possibly zero-probability cells
• Relations among conditional probabilities
• Algebraic causality: Bayes nets and beyond
• Markov bases of binary graph models of K_4-minor free graphs
• Graphical models for correlated defaults
• On the Geometry of Discrete Exponential Families with Application to Exponential Random Graph Models
• A class of statistical models to weaken independence in two-way contingency tables
• Polynomial relations among principal minors of a 4x4-matrix
• Normal binary graph models
• Graphical methods for efficient likelihood inference in Gaussian covariance models
• Specializations of multigradings and the arithmetical rank of lattice ideals