I think the answer is Yes and No and both
are correct.
The syntax for -bootstrap- makes it clear
that it is a framework for repeatedly
running a single command.
However, there is nothing to stop
you putting a sequence of commands in
a single program. To Stata that is then
just another command. The key is that the
results must be visible to -bootstrap-.
(At a guess, this would work with
a do file too, with the same caveat.)
Nick
[email protected]
Scott Imberman
> I would like to know if there is any way to bootstrap over multiple
> commands in Stata9. In particular i'm trying to bootstrap
> over matching
> estimators using psmatch2. However, I have too many variables in my
> propenstity score for psmatch2 to use directly and i need to
> weight the
> propensity scores which cannot be done in psmatch2. Thus, to get my
> bootstrap standard errors i need to bootstrap over a probit command, a
> prediction of the propensity score, and then psmatch2. I'd greatly
> appreciate any help you can give.
*
* 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/
