Somsupa Nopprach wrote:
I wonder that is there any different in running
random coefficient logit model by using -xtlogit- and
using -gllamm- command?
--------------------------------------------------------------------------------
Roger Harbord and Alexander Staus have already given some advice about the
greater generality of -gllamm-. Let me just add that, when you're working
with these nonlinear models, you'll want to use *adaptive* Gauss-Hermite
quadrature. -xtlogit- and -gllamm- use different implementations of
adaptive Gauss-Hermite quadrature. As I recall (the posts are on the List
within the past year), -gllamm-'s implementation is somewhat slower but more
accurate than -xtlogit-'s. StataCorp was looking into this matter at last
report.
Joseph Coveney
*
* 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/
