Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: subpopulations, subpopulation weights in gllamm


From   Stas Kolenikov <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: subpopulations, subpopulation weights in gllamm
Date   Mon, 31 Oct 2011 08:00:53 -0500

-gllamm- does not support -subpop()- option, as the latter is the
option of -svy- only. Specifying -if- might work OK in your case if
you find overweight children in every cluster (in other words, if your
subpopulation does not eliminate any clusters). I would suggest
looking at West, Berglund and Heeringa (2008,
http://stata-journal.com/article.html?article=st0153). They identify
this and other situations when specifying an -if- condition is
equivalent to specifying -subpop()- option (although generally, of
course, -if- is an absolute no-no with survey data, as you are well
aware of). Another way of specifying the subpopulation is to manually
zero out the weights outside of subpopulation. You could try different
approaches and see if they produce the same answer.

On Sat, Oct 29, 2011 at 12:41 PM, Lisa Marie Yarnell
<lisayarnell@yahoo.com> wrote:
> As a follow-up to my question, it is important to preserve the validity of pweight when I run these subpopulation models.  If I am interested in running the model for the subpopulation of overweight youth in the sample, how do I specify that I want to run the model only for that subpopulation (but preserve the validity of the sampling weight)?
>
> Thanks!
>
>
> --- On Sat, 10/29/11, Lisa Marie Yarnell <lisayarnell@yahoo.com> wrote:
>
>> From: Lisa Marie Yarnell <lisayarnell@yahoo.com>
>> Subject: st: subpopulations, subpopulation weights in gllamm
>> To: statalist@hsphsun2.harvard.edu
>> Date: Saturday, October 29, 2011, 4:39 AM
>> Hello!  I estimated some models
>> using gllamm that had persons nested in schools, and had
>> grand sample weights applied.  For example, I ran this
>> model:
>> gllamm notdep bmi boy aa hispanic ses, i(schoolid)
>> family(binom) link(logit) eform pweight(wt)
>>
>>
>> If I want to run the model for subgroups such as overweight
>> / not overweight, is it sufficient to simply type "if
>> overweight = 1" in the model, like this:
>> gllamm notdep bmi boy aa hispanic ses, i(schoolid)
>> family(binom) link(logit) eform pweight(wt) if overweight =
>> 1
>>
>>
>> OR do I need to put the subgroup somewhere else in the
>> model, as in some versions of Stata outside of the gllamm
>> module, such as:
>> gllamm(overweight) notdep bmi boy aa hispanic ses,
>> i(schoolid) family(binom) link(logit) eform pweight(wt)
>>
>>
>> I would like gllamm to use proper subpopulation weighting
>> when I select the overweight group.  I would be really
>> appreciative of any help I can get with this.  Thank
>> you for your time.
>
>
> *
> *   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/
>



-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   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   |   Site index