Thank you very much for your response. Sorry, I was not specific
enough!
I estimate the following model:
eoprob y x1 x2 x3 x4, i (id)
thereafter, I would like to get the marg. effects for some
variables x1 and x2 for outcome 1 (e.g.) and use the post
estimation command
mfx, predict (p outcome(1)) varlist(x1 x2)
this however, is not supported by reoporb. Has somebody already
coded that?
Thanks a lot and best,
Mark
----Ursprüngliche Nachricht----
Von: [email protected]
Datum: 10.10.2006 23:59
An: <[email protected]>
Betreff: Re: st: marginal effects after random effects ordered
probit
On Oct 10, 2006, at 5:53 PM, [email protected] wrote:
> Dear all
>
> Has anyone coded the mfx function to work after the random
effects
> ordered probit estimator (reoprob)? Guillaume Frechette, the
author
> of reoprob, said that he had many inquiries for the mfx
> postestimation function, and somebody might already have coded
it.
>
> Thanks a lot,
> Mark
> *
> * 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/
I'm assuming that the straightforward syntax of -mfx- doesn't
work
then with -reoprob-? With nonlinear models, I usually do
something
like:
- mfx, dydx-
- mfx, eyex-
with the first giving me the coefficients in terms of level on
level
and the latter in terms of an elasticity. Your question makes
me
think -reoprob- may be different, but that's how you'd do it
with,
say, poisson or nbreg.
*
* 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/
