Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: p-values and -test- when bootstrapping svyreg


From   jpitblado@stata.com (Jeff Pitblado, StataCorp LP)
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: p-values and -test- when bootstrapping svyreg
Date   Wed, 05 May 2004 17:06:11 -0500

Cecilia von Otter <cecilia.vonotter@sofi.su.se> asks about using -svyreg- with
-bootstrap-:

> We're bootstrapping svyreg (using pweights).
> Why does 'test' of  the bootstrapped coefficents run chi-square-tests ?
> What happened to old adjusted Wald f-tests?
> Is it safe to calculate t-stats on _b/ s.e or do I need to specify that
> I want the bootstrap procedure to retrieve t-stats?
> Can I ask for p-values when bootstrapping svyreg?

The -bootstrap- command does not perform sampling that takes weighted
observations into account, nor can it handle multiple stages of cluster
sampling.

For these reasons, the -bootstrap- command is generally not recommended for
use with survey data analysis; except for the cases where there is at most one
stage of clustering and the weights are constant within stratum.

Even for these cases, the -bootstrap- command does not post a "residual"
degrees of freedom so that -test- will perform an F test instead of a
Chi-Square test.

--Jeff
jpitblado@stata.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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index