There is a (in my opinion not very populated) school of thought that says
that the post-stratification and raking of weights changes variance
estimates. I am only aware of WesVar specifically addressing this issue.
If you accept that the scale of the weights does not have much (if any)
impact on the estimates, then the standard -svy- commands in Stata work
fine. The advantage of post-stratification is that you get totals that have
meaning, like the number of people unemployed.
Bryan Sayer
Statistician, SSS Inc.
-----Original Message-----
From: Henrique Neder
To: [email protected]
Sent: 9/17/02 4:09 PM
Subject: st: post-stratification
I corrected the message because it had made a mistake when referring to
post-estimation. In reality it is post-stratification. The right message
is:
Did anybody already work with post-stratification in complex survey
samples (cluster sampling) using Stata or know some work about this? I
have some data - a household survey using two-stage sampling with
weights and "post-stratification factors" (I don�t know if this is the
correct name), the identificator of PSUs and Strata and I want to
estimate confidence intervals for poverty indicators. The survey (svy)
Stata doesn't have sub-commands to consider post-stratification.
Thanks in advance
Henrique Neder
Instituto de Economia
Universidade Federal de Uberl�ndia
Minas Gerais - Brazil
[email protected]
*
* 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/
