Statalist The Stata Listserver


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

Re: st: Zero-inflated Poisson/Negative Binomial for survey data


From   jpitblado@stata.com (Jeff Pitblado, StataCorp LP)
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Zero-inflated Poisson/Negative Binomial for survey data
Date   Mon, 17 Apr 2006 16:12:22 -0500

Yu, Binbing (IMS) <YuB@imsweb.com> wants to know if -svy- is supported by the
-zip- and -zinb- commands:

> I know that STATA can fit various regression to complex survey data as well
> as zero-inflated Poisson (ZIP) and negative binomial (ZINB) for simple
> random sample (SRS).  I wonder if STATA can do ZIP/ZINP for complex survey
> data?

The following is true for Stata 8 and Stata 9, although Stata 8 requires an
extra -score()- option in the call to -zip- (and -zinb-).

While -zip- and -zinb- do not yet have a counterpart for survey data analysis,
you can use -suest- with the -svy- option to get linearized variance estimates
according to the design characteristics in your -svyset-.

For example say the sampling weight variable is -sampwgt-, then you can do
something like:

	. svyset ... [pweight=sampwgt], ...
	. zip ... [iweight=sampwgt], ...
	. suest ., svy

The name of the 'pweight' variable in -svyset- must match the 'iweight'
variable used in -zip- (and -zinb-).

--Jeff
jpitblado@stata.com
*
*   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