st: Problems fitting twin data (ACE model) with sem command

[Jacob Felson <felsonj@gmail.com>]

I'm wondering if it is possible to estimate a basic ACE (classic twin
model) using the sem command.  I think such a model would require
either that error terms be dropped.  It doesn't seem possible to drop
the error terms
for endogenous variables in sem.

Is there a workaround?

When I tried to fit such a model, Stata immediately produced an error: the
model was unidentified; too many latent variables.



Jacob Felson


--
Jacob Felson
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/