Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: RE: st: Strange behavior or bug? - svy : total returns different


From   jpitblado@stata.com (Jeff Pitblado, StataCorp LP)
To   statalist@hsphsun2.harvard.edu
Subject   Re: RE: st: Strange behavior or bug? - svy : total returns different
Date   Fri, 28 Aug 2009 17:40:23 -0500

Zurab Sajaia <zsajaia@hotmail.com> has a few follow-up questions about missing
values an subpopulation estimation:

> Thanks a lot for the explanation, although I wouldn't say that I got the
> point - if zero weight-observations are interpreted as "out of interest" why
> are thy kept in the sample? and if they are, then why aren't they dropped as
> containing missing values in the variable of interest? (mv)

Zurab is using -svy: total-, thus the rules for subpopulation estimation
apply.

Observations outside the subpopulation are not used to compute the point
estimates but they are accounted for in the variance calculation.  Thus they
are still considered part of the estimation sample.

Since Stata 10, the -svy- prefix keeps out of subpop observations in the
estimation sample regardless of missing value pattern in the model variables
so that it can properly perform the subpopulation variance estimation.

--Jeff
jpitblado@stata.com
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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