st: biprobit predict command

[Jean-Christophe Boucher <Jean-Christophe.Boucher@hei.ulaval.ca>]

Hi,
I am using a bivariate probit model with partial observability (biprobit)
where some of the IVs in the first equation are either binary or continuous.
Right now I have used the adjust stata command in order to predict
estimations of the continuous IVs while keeping all other IVs at their
means. There are two issues that worries me:
1) Is keeping binary IVs at their means to estimate predictions is worrisome
2) In bivariate models, the prediction commands only works for the first
equations, should I be worried about not accounting for the rho in the
prediction? Or is it already included in the biprobit results? Is there any
specific command to use post-estimation technics for multilevel model such
as the one I am using?

JC
*
*   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/