very conservative bootstrap proportions

[Jan Bosselaers]

Dear all,

I would like to ask a small technical question to the list, perhaps of
too limited general interest, but I try nevertheless.
Philip Chu (1998) Cladistics 14, p.8 states that, when the number of
searches allowed in each iteration of a bootstrap analysis of a data
matrix for cladistic analysis is very limited, bootstrap proportions
tend to be too conservative: "with the former (meant is the bootstrap),
if non-minimal trees are found in each pseudoreplicate, then the
bootstrap proportions for each group may be lower than if minimal trees
had been found"
My question: is the same true for the jackknife? I would guess so, since
"In other words the jackknife is, almost, a bootstrap itself" (Efron and
Gong (1983) The American Statistician 37(1): 36-48 (p. 40)), but I must
humbly admit that I am not entirely sure. Can anybody advise?

--------------------------------------------------------------
Jan Bosselaers
"Dochterland", R. novarumlaan 2
B-2340 Beerse, Belgium                       tel 32-14-615896
home: dochterland at village.uunet.be           fax 32-14-610306
work: jbossela at janbe.jnj.com
web: http://gallery.uunet.be/Dochterland/
"Non intellegunt homines, quam magnum vectigal sit parsimonia"
Cicero