Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Bootstrapping weighted glm regression
From
Stas Kolenikov <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: Bootstrapping weighted glm regression
Date
Wed, 21 Aug 2013 08:48:40 -0500
I demonstrated how to conduct weighted bootstrap outside of the
complex survey context in my 2010 SJ paper, which is probably freely
available now: http://www.stata-journal.com/article.html?article=st0187.
-- Stas Kolenikov, PhD, PStat (ASA, SSC)
-- Senior Survey Statistician, Abt SRBI
-- Opinions stated in this email are mine only, and do not reflect the
position of my employer
-- http://stas.kolenikov.name
On Wed, Aug 21, 2013 at 8:07 AM, Stephen Amrock
<[email protected]> wrote:
> Hi,
>
> I'm conducting survival analysis using inverse probability weighting
> (IPW) (i.e., stabilized weights generated from propensity scores). I'm
> hopeful someone will be able to help with some guidance regarding
> Stata's (1) weighting; and (2) bootstrap features. It appears that
> bootstrap never can work with weighted data. Is there a workaround for
> this?
>
> Stata, for example, can compute:
>
> glm [outcome] [treatment], family(binomial) link(log) eform vce(boot,
> r(1000) nodots)
>
> but can NOT perform bootstrapping for the following:
>
> glm [outcome] [treatment] [pw=weight], family(binomial) link(log)
> eform vce(boot, r(1000) nodots)
>
> Any advice would be much appreciated, since Statalist archives I've
> found deal with complex survey sampling (i.e., BRR).
>
> Thanks and best wishes,
> Steve
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/