If you include the stratification in your model, and specify the PSU as a
cluster variable, you can 'approximate' the results. Just how close you
get, I don't know. I believe you would need to interact the stratification
with all of the rest of the variables to fully account for it. Now of
course, your model looks different so I don't know just how one would make a
comparison.
Bryan Sayer
Statistician, SSS Inc.
-----Original Message-----
From: Javier Escobal
To: [email protected]
Sent: 12/11/02 7:41 PM
Subject: st: svyreg versus regress
Dear statalisters
Can anybody tell me how close I can get using regress to mimic the
results provided by svyreg? How would the instruction look like? Is
there anyway I can reproduce the same standard errors if I have an
stratified sample?
regards
Javier
*
* 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/
