Dear colleagues:
I have revisited with some students the paper by Pearl and Wermuth
entitled "When Can Association Graphs Admit a Causal Interpretation?"
There are several versions of this paper (in the Fourth Workshop on AI and
Statistics, 1993, a Springer book that came out of that workshop, and
technical report R-183 at http://bayes.cs.ucla.edu/jp_home.html). I refer
here to the R-183 version, which corrects some errors in the AIS workshop
version.
My renewed interest in this work is due to the fact that I have some large
correlation matrices, and I would like to see if the corresponding
independence structure can be explained by a DAG model. The 1993 paper
seems to address precisely this problem.
I wonder whether you have some suggestions for related later work on the
same topic. I know of Verma and Pearl's "Morality of Graphs is
NP-Complete" (UAI-93). Anything else?
Your comments will be most appreciated.
Best regards,
Marco
Marco Valtorta Associate Professor
Department of Computer Science and Engineering
University of South Carolina
Columbia, SC 29208, U.S.A. http://www.cse.sc.edu/~mgv/
803.777-4641 fax:-3767 [EMAIL PROTECTED]
