Question Regarding Bootstrap
hoelzer at unr.edu
Fri Oct 29 12:10:38 EST 1999
In article <7va4cs$rkr at net.bio.net>, James McInerney
<james.o.mcinerney at may.ie> wrote:
> A Mitchell wrote:
> > Yes I understand your point about variance, but there is a paper in the
> > literature (Bull et al.?) suggesting that 500 bootstrap replicates is
> > sufficient for most purposes. One million just seems way out there...
First, you need to define what constitutes a sufficient number for
yourself. Usually, it means something like the bootstrap value does not
change much if you continue to increase the number of replicates.
However, the degree of change is a continuous thing, so there is no black
and white threshold there. Also, the results for Bull et al. were based
on very particular data sets (was there more than one?). The way variance
in bootstrap values settle down with increasing numbers of replicates is
data set specific; so, generalizing the results from Bull et al. to your
data can be misleading.
Department of Biology
University of Nevada Reno
Reno, NV 89557
More information about the Mol-evol