Hi everybody!
Are there any of us who carry out bootstrapping during bayesian inference? I would be interested if this is a good idea and what can we expect from it?
I met this approach in Douady & al Mol Biol Evol. 20(2): 248, 2003., and (to me at least) seems to be an intuitively easily understandable alternative for fighting with overestimation by posterior probabilites. I can understand the latter as if one would use the trees produced by a normal heuristic search as a base for clade credibility values. In fact nobody does this in "normal" ML and MP searches, but why should we do so in Bayesian analyses?
Thanks,
Laszlo
---------------------------------
Yahoo! Messenger - with free PC-PC calling and photo sharing.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.bio.net/bionet/mm/mol-evol/attachments/20061205/a2f3abf7/attachment.html
