Re: st: post estimation in gllamm

[n j cox <n.j.cox@durham.ac.uk>]

In principle, I suppose, sending to the list and lodging a copy in
the archives are different. In practice, however, the only evidence
accessible to all on what has been sent is that in the
archives. The fact that any individual may not have got any
particular posting is presumably attributable to local circumstances, 
whether known or unknown. Thus, if something is the archives it
should be assumed that it was sent and received.

I guess that the reason you did not get a reply is that
the question is well nigh impossible to answer without
more information on your data and your model command.
The Statalist FAQ does ask you to give precise information on what
you typed. That said, I have no idea what kind of information a -gllamm-
expert would need here. But in general your number of observations
may be much more than a command may be able to use.

See also Statalist FAQ advice on repeat postings. Changing a title
does no harm if you think that your previous title was misleading
but by itself it doesn't render an unanswerable question answerable.

Nick
n.j.cox@durham.ac.uk

Michele Guerin wrote

This is a repeat of an earlier e-mail but with a new title.  I haven't
received a response yet and wasn't sure if the mail was posted to the
listserve (although the original message was in the archives)

I am having a problem with post-estimation in gllamm.  I am using Stata
SE version 9.2.  I have a dichotomous outcome, and used the adaptive
quadrature option.

When I use any of the gllapred commands, the error message says that I
have "insufficient observations".  My dataset contains:

Level 1:  694 observations
Level 2:  80 observations
Level 3:  31 observations

Can anyone offer a reason for this and what I might do about it?
*
*   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/