Dongyi Du <[email protected]> asks how to generate bootstrap replicate weights:
> I have information of weight, psu, and strata for a survey data, can
> anyone tell me if I can generate my own bootstrap weights? if I can,
> how?
There doesn't seem to be much consensus in the survey literature regarding the
bootstrap.
Stas Kolenikov <[email protected]> suggested a paper by Rao & Wu (giving a
full reference) and gives some advice. Although I cannot suggest how to
generate your "replicate weight variables", I can point you to my unofficial
-bs4rw- command once you have figured a way of generating them. For
reference, here is the archive of my original Statalist post about -bs4rw-
http://www.stata.com/statalist/archive/2005-09/msg00029.html
Dongyi did not indicate the purpose/need for the 'bootstrap weights', but
maybe -svy jackknife- (or -svy brr- assuming a 2 PSU/Stratum design) can be
used for variance estimation.
--Jeff
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
