On 7 Feb 2000, Joe Felsenstein wrote:
> From Quenoville's reported results, the branch and bound for ML in PAUP* is
> not doing as well as exhaustive search, and even takes longer!
That's indeed a bit surprising because the worst-case behaviour of
branch and bound should *is* exhaustive search ..
> The reason no one has written a paper on branch and bound for ML is that
> we don't know a good way to compute a tight bound.
Yes, I agree that finding such an estimate is the difficulty here.
Thanks also for confirming the non-existence of a b&b ML paper.
Korbinian Strimmer http://users.ox.ac.uk/~strimmer
Dept. of Zoology, Univ. of Oxford, South Parks Road, Oxford OX1 3PS
+44 1865 271272 (phone), -49 (fax), korbinian.strimmer at zoo.ox.ac.uk