Garry,
If the intraclass correlation is very high and/or
your panels are large, you may be encountering
numerical problems. In these cases, adaptive
quadrature may work better which is available
in gllamm (see Stata Journal paper, volume 2,
no.1 p.1-21)
Good luck!
Sophia
----- Original Message -----
From: "Garry Anderson" <[email protected]>
To: <[email protected]>
Sent: Wednesday, May 28, 2003 11:27 AM
Subject: st: dot for standard error in xtlogit
> Hi,
>
> I am receiving a dot for the standard error that relates to a coefficient
> in xtlogit
> Panels with zero positives do not seem to be the problem as other
> estimations with xtlogit that do have some panels with zero positives do
> not have a dot in the results.
>
> Kind regards, Garry
> [email protected]
>
> *
> * 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/
*
* 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/
