Statalist The Stata Listserver


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

st: RE: weighted least squares with fixed effects


From   "Steve Stillman" <stillman@motu.org.nz>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: weighted least squares with fixed effects
Date   Wed, 17 May 2006 02:53:26 +1200

Just guessing from your variable names but I believe you want to use aweights or pweights not fweights.  fweights are frequency weights, they are meant to do exactly what you report, eg. treat each observation like it represents X actual observations.  aweights should be used if you are trying to account for difference in sampling variation usually caused by the regression variables being sample averages or pweights if you are accounting for population-weighted sampling.

BTW, this information is all in the stata manual.

Cheers,
Steve 

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu]On Behalf Of Allers
Sent: Tuesday, May 16, 2006 11:49 PM
To: statalist@hsphsun2.harvard.edu
Subject: st: weighted least squares with fixed effects


Dear Stata users,

The -areg- command can estimate a fixed effects model with weights,
e.g.: areg  depvar indepvars [fweight=population], absorb(countycode).
However, when I do this, I get standard errors that are way too low,
because Stata artificially increases the sample size with the weight
variable (here: population). Adding -robust- is no solution. Does anyone
have a solution?

Regards,

Maarten

*
*   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/


*
*   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