RE: st: survey bootstrap

[Jun Xu <mystata@hotmail.com>]

Steve,

Thanks. I have make a series of group comparison, so I use bootstrap to correct for family-wise alpha level.

----------------------------------------
> From: sjsamuels@gmail.com
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: survey bootstrap
> Date: Sun, 6 Feb 2011 00:26:13 -0500
>
> You can do it with Stas Kolenikov's -bsweights- command ("findit").
> Be sure to read his presentation http://www.stata.com/meeting/snasug08/kolenikov_snasug08.pdf
> . I'm curious: for your problem, why do you prefer the bootstrap
> over linearization?
>
> Steve
> sjsamuels@gmail.com
>
>
>
> On Feb 5, 2011, at 2:58 PM, Jun Xu wrote:
>
> Dear Stata Community,
>
> I have a data file with complex survey design. While setting up my
> svy structure, I have
>
> svyset cluster [pw = var1], strata(var2)
> svy: logit y x1 x2 x3
>
> What
> I want to do is to bootstrap standard errors for x1, x2 and x3 instead
> of not the linearized s.e. produced by the svy command. I looked at the
> svy bootstrap command help file, but I don't quite get how to create
> replicate weight variables required by the bsrweights option. I can go
> one step further to use bsample, but it appears that its options
> (strata, cluster, weight) do not accept pweight.... Any suggestion?
>
> Jun Xu, PhD
> Assistant Professor
> Department of Sociology
> Ball State University
> Muncie, IN 47306
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
 		 	   		  
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/