You can try:
1) user-written command -mfx2- from http://fmwww.bc.edu/RePEc/bocode/m extending official command -mfx- (I tried and seemed to work, not sure about how much "good" std. errors are)
2) user-written command -divprob- from http://fmwww.bc.edu/RePEc/bocode/i implementing Amemiya Generalized Least Squares (AGLS) estimators (no robust std. errors) --- if you don't find this one, write me off the list and I will send it to you
Please don't ask me what AGLS is.
Nicola
At 02.33 26/01/2007 -0500, Matt D Mitchell wrote:
>Hello,
>
>I am trying to obtain marginal effects on ivprobit.
>
>As I understand it, I need to use the post-estimation command mfx, predict(p). "mfx" alone will not work (it does not give the marginal effects, I believe). When I use the predict(p) command, however, Stata reports that the expression is unsuitable for calculating standard errors. I know I can force it, but Stata says this may not be recommended. Is it okay to use force if the relative difference is very small?
>
>Any recommendations on how I can obtain the marginal effects of an ivprobit and good standard errors?
>
>Much appreciated,
>
>Matt
*
* 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/
