Would you have to specify the weights? The -svy- commands use
linearization to obtain appropriate standard errors. Bootstrapping is
an alternative method although Balanced Repeated Replication and
Jackknifing are more commonly mentioned techniques. So it's not clear
to me what -svybs- can do that -bs- with the cluster option cannot.
You might want to take a look at "Lohr, S.L. (1999). Sampling: Design
and Analysis. Pacific Grove: Duxbury Press." She discusses
bootstrapping in relationship to complex surveys, IIRC.
Good luck,
John Hendrickx
--- "Shinogle, Judy" <[email protected]> wrote:
> I have a complex survey data and need to bootstrap an estimate
> taking into account PSU, strata and weights. I found stans svybs
> that takes into account PSU but what about weights. My first
> thought was to expand the data by the weight but that would make
> the dataset too large for my memory >50mill. Any ideas?
__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com
*
* 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/
