On finite exchangeability and conditional independence (Q2192320)

The main result of this article is the following: Theorem. If a distribution \(P\) over an exchangeable random network with \(n\) nodes is faithful to a chain mixed graph \(G\), then \(G\) is Markov equivalent to one of the following graphs: the empty graph, the undirected incidence, the bidirected incidence graph, the undirected complement of the incidence graph, the bidirected complement of the incidence graph, the complete graph.