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: GLLAMM versus XTMEPOISSON


From   Ana Cecilia Montes Vinas <ac.montes393@uniandes.edu.co>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: GLLAMM versus XTMEPOISSON
Date   Tue, 12 Feb 2013 18:50:46 +0000

Thank you. 
Im gonna try that 

-----Mensaje original-----
De: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] En nombre de Stas Kolenikov
Enviado el: martes, 12 de febrero de 2013 12:04 p.m.
Para: statalist@hsphsun2.harvard.edu
Asunto: Re: st: GLLAMM versus XTMEPOISSON

Ana,

these are the parameters of the Cholesky decomposition of the variance-covariance matrix. The lns are the natural logs of the diagonal elements, and atr is the (hyperbolic?) arctan of the correlation. -gllamm- produces appropriately transformed coefficients in the output, but I don't think the matrix of the random effect appears in the output or saved -e()- results.

Try running -xtmepoisson- using -gllamm-'s results as starting values, and vice versa. May be you are finding local optima. Also, try changing the number of integration points, that's a crucial parameter in numeric integration of mixed models.

--
-- Stas Kolenikov, PhD, PStat (SSC)  ::  http://stas.kolenikov.name
-- Senior Survey Statistician, Abt SRBI  ::  work email kolenikovs at srbi dot com
-- Opinions stated in this email are mine only, and do not reflect the position of my employer


On Tue, Feb 12, 2013 at 9:48 AM, Ana Cecilia Montes Vinas
<ac.montes393@uniandes.edu.co> wrote:
> Dear statalisters
>
> I'm estimating a Multilevel poisson regression using xtmepoisson and gllamm  commands. I have two types of characteristics, firm individual characteristics (x1) and sector characteristics (x2). However, when we use gllamm and Xtemepoisson we obtain contradictory results, in particular, x1 is negative with xtmepoisson, and positive  with gllamm.
>
> eq ri: cons
> eq rc: x1
> matrix a = e(b)
> gllamm y x1 x2, family(poisson) link(log) i(ciiu) nrf(2) eqs(ri  rc) from(a) eform adapt
>
>
> xtmepoisson y x1 x2 || ciiu:x1, irr cov(unstructured)
>
> Adicionally when i extract the e(b) matrix, i get 3 things called lns1_1_1, lns1_1_2, atr1_1_1_2, and i'm not sure if they are the variances and covariaces of the random effects.
>
> Thank you
>
> Ana C
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index