Mau et al. max likelihood article

Richard Zander bryo at COMMTECH.NET
Fri Sep 5 10:02:30 CDT 1997


I quoted the article on maximum likelihood wrong in my recent post.
Correctly, I hope, it is:
http://www.stat.wisc.edu/~newton/papers/abstracts/tr96a.html

This is an important paper, being one of the first that uses modern
methods of max likelihood analysis on large numbers of terminal taxa.
The posterior probabilities of the 31-taxon cichlid fish study, of which
the highest is only .11, are given as frequencies in the tables. This
means in the frequentist sense that the tree of maximum likelihood (of
explaining the data set) would be expected to be correct in about 1 out
of 10 occurrences of this same data set in the long run. Assuming all
the many regularity assumptions are correct....

--

*******************************************************
Richard H. Zander, Buffalo Museum of Science
1020 Humboldt Pkwy, Buffalo, NY 14211 USA bryo at commtech.net
*******************************************************




More information about the Taxacom mailing list