Thomas Corneli�en asks about fitting bivariate ordered probit models and
notes an earlier thread mentioning -gllamm-. The user's manual for -gllamm-
contains a citation to Leonardo Grilli & Carla Rampichini, Alternative
specifications of bivariate [multivariate] multilevel probit ordinal
response models. _Journal of Educational and Behavioral Statistics_
28:31-44, 2003. A related working-paper version is available for
downloading at
www.ds.unifi.it/ricerca/pubblicazioni/working_papers/2002/wp2002_08.pdf
To get rho after -gllamm-, divide the random effects variance by the sum of
the residual variance (which is 1.0 for probit) and the random effects
variance. If the random effect is named pid as in the do-file shown, then
it would be
display _b[pid1:_cons]^2 / (1 + _b[pid1:_cons]^2)
Joseph Coveney
*
* 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/
