Sep. 13 2005 2:00PM

Sep. 13 2005 3:30PM

souhrn:
Chain graphs (= adicyclic graphs) have both undirected and directed edges and can be used to represent simultaneously both structural and associative dependences. Like acyclic directed graphs (ADGs = DAGs), the chain graph associated with a given statistical model may not be unique, so chain graphs fall into Markov equivalence classes, which may be super-exponentially large, leading to possible ambiguity and computational inefficiency in model search and selection. It is shown here that under the Andersson-Madigan-Perlman (AMP) Markov interpretation of a chain graph, each Markov-equivalence class can be uniquely represented by a single chain graph, the AMP essential graph that is itself simultaneously Markov equivalent to all chain graphs in the AMP Markov equivalence class. A complete characterization of AMP essential graphs is obtained.

Pravidelný seminář Inteligentní systémy (Kontaktní osoba: Tomáš Kroupa)