Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Pweights with random effect model


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Pweights with random effect model
Date   Wed, 1 Nov 2006 13:59:09 -0500

Michael and Sara--
The point estimates obtained using [aweights] and [pweights] are in
most cases identical, but the standard errors are not.  In short,
[pweights]=[aweights]+_robust (see -help _robust-). Sometimes you can
use aweights and bootstrap standard errors when pweights are not
allowed. See also:
http://www.stata.com/support/faqs/stat/supweight.html

However, -xtreg- with the -re- option allows no weights, and with the
-mle-option takes only [iweights].  It may be possible to specify your
pweights as [iw=pweights] but inference is a trickier business, I
suspect.

On 10/31/06, Michael.D.Wilson@frb.gov <Michael.D.Wilson@frb.gov> wrote:
I dont know if this is econometrically sound, but I've used aweights with
a random effects model and haven't had a problem.  Several times I've
compared results using aweights and pweights and have not found much difference if any.

"Sara Bleich" <bleich@fas.harvard.edu>
Does anyone know of a command that allows you to use pweights with a random effect model? Xtreg does not, but is there another command that does?
*
*   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