Hello everyone,
my questionis related to the article of Norton, Ai and Wang (Computing interaction effects and standard errors in logit and probits models), The Stata journal, 2004, 4(2).
In this paper, the authors explain "why computing the marginal effect of a change of two variables is more complicated in nonlinear models (probit, logit, tobit) than in linear models". That is to say, the Stata command mfx does not give a good marginal effect when it is an interaction effect. The command "inteff", which the authors drecribed in this paper, allows to compute a good marginal effect of a change of two variables for probit and logit models, but not for the tobit.
I need one for the tobit. I read the "inteff.ado" file. Is it possible to change a little bit the "inteff.ado" file and then get the program for computing the marginal effect of a change of two variables for a tobit model (with ramdom effect).
I would like to do myself. Unfornately, I am neophyte, so I do not even understand the programm of the "inteff.ado" file.
Thank you
Etienne Dumont
Université Laval (Canada)
*
* 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/
