On Tue, Nov 24, 2009 at 2:08 PM, Roberto G. Gutierrez, StataCorp
<[email protected]> wrote:
> Such a covariance is, of course,
> unidentified because a child cannot be both male and female within the course
> of this study.
... and that explains missing standard errors in -xtmixed- output, as
well as correlations of exactly 1 in -gllamm- output.
I wonder what Mplus does there. They probably just set it to zero,
that may be the most sensible thing to do.
--
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.
*
* 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/
