Re: st: different standard errors with gllamm vs. xtmelogit

[rgutierrez@stata.com (Roberto G. Gutierrez, StataCorp)]

Robert de Vries, Robert <r.de-vries08@imperial.ac.uk> is having some
difficulty reconciling results between -gllamm- and -xtmelogit-.  He writes:

> The log likelihood's with 7 integration points are:

> Xtmelogit: -15047.008
> Gllamm:  -15056.655

> I tried increasing the number of integration points to 15 as you suggested.
> This yielded yet more strange results.

> The gllamm and xtmelogit likelihood's are more similar for these models, but
> with gllamm's still being slightly higher (15047.916 vs. 15059.682)

> The standard errors of the coefficient I'm interested in are still wildly
> different between gllamm and xtmelogit, but now so is the coefficient
> itself:

> Xtmelogit: -.034(.039663)
> Gllamm: -.0045(.004751)

> Most of the other coefficients and standard errors are almost identical
> between the two models so initially I thought it might just be a problem
> with the level 2 variable I've been discussing above. However the
> coefficients and SE's of the individual l evel binary gender variable are
> also very different between gllamm and xtmelogit.

In a previous message in this thread Robert showed an example of a -gllamm-
command that he issued, and this command did not include the -adapt- option.
-adapt- tells -gllamm- to use adaptive rather than standard Gaussian
quadrature.  Because -xtmelogit- also uses adaptive Gaussian quadrature
(albeit in a slightly different form), using -adapt- with -gllamm- should help
bring the results more in line with one another.

Of course, if the difference persists Robert can feel free to email me with
the data and I can take a closer look.

--Bobby
rgutierrez@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/