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: Question regarding GLLAMM


From   Stas Kolenikov <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Question regarding GLLAMM
Date   Fri, 13 Aug 2010 19:20:44 +0400

It will work fine, although with 12000 level-1 data points times 14
"level-0" responses = 170K effective GLLAMM observations and three
levels, it will probably take some three days to converge on an
average computer. You probably want to fit some sort of a MIMIC model
where you have three or so latent factors addressed by your 14 items;
and these factors are regressed on the demographic variables. I have
run something very similar with NHANES data, although at two levels
(items-patients).

On Thu, Aug 12, 2010 at 1:55 AM, Shubhabrata Mukherjee
<joy_stat@yahoo.com> wrote:
> Hi,
>
> I want to fit a hierarchical model and was wondering if GLLAMM can be used to
> fit the model. My data loos like this:
>
>    I have item responses (polytomous) from almost 12000 'patients'). They
> answer 14 questions (I1-I14) where some of the questions
> are regarding their doctors and some regarding the clinic as a whole. I have
> three variables (characteristics) for patients:
> ethnicity (white, black and hispanic), education (dichotomized) and age
> (continous) which I want to embed in the model.
>
>  The number of doctors are 400. The data I have is cross-sectional. One patient
> is designated to a single doctor only.
>
>   In the next level, the doctors fall under 8 different clinics. Again a doctor
> cannot be in more than one clinic.
>
>   So I have this hierarchy: item responses --> patients --> doctors -->
> clinics.
>
>   In my case I have 14 response variables instead of one and I have this
> hierarchy. Can GLLAMM fit the following model? or will it
> be too complicated?
>
>
> Thanks,
>
>  -Joey
>
>
>
>
> *
> *   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