Title | ||
---|---|---|
A Transformational Characterization of Markov Equivalence for Directed Acyclic Graphs with Latent Variables |
Abstract | ||
---|---|---|
Difierent directed acyclic graphs (DAGs) may be Markov equivalent in the sense that they entail the same conditional indepen- dence relations among the observed vari- ables. Chickering (1995) provided a transfor- mational characterization of Markov equiv- alence for DAGs (with no latent variables), which is useful in deriving properties shared by Markov equivalent DAGs, and, with cer- tain generalization, is needed to prove the as- ymptotic correctness of a search procedure over Markov equivalence classes, known as the GES algorithm. For DAG models with latent variables, max- imal ancestral graphs (MAGs) provide a neat representation that facilitates model search. However, no transformational char- acterization | analogous to Chickering's | of Markov equivalent MAGs is yet available. This paper establishes such a characteriza- tion for directed MAGs, which we expect will have similar uses as it does for DAGs. |
Year | Venue | Keywords |
---|---|---|
2005 | Uncertainty in Artificial Intelligence | latent variable,directed acyclic graph |
DocType | Volume | Citations |
Conference | abs/1207.1419 | 4 |
PageRank | References | Authors |
0.60 | 2 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jiji Zhang | 1 | 149 | 17.52 |
Peter Spirtes | 2 | 616 | 101.07 |