Re: st: marginal effects after TPM-Poisson model (or hurdle model)

[May Boggess <mboggess@stata.com>]

On Thursday morning, Sandra asked two questions about marginal effects:

>  How I can get the marginal effects of the
> second equation? 
> Is there a way to estimate marginal effects of the
> logit equation predicting probabilities?
The answer to both questions lies in the -predict()- option on -mfx-. It 
is used to specify the function of which -mfx- will find the slope. It 
can find the slope of any function that -predict- can  calculate after 
the precedding estimation command. So the key to getting -mfx- to do 
what you want is understanding what -predict- can calculate and then
using the -predict()- option on -mfx- to specify the prediction you need.

Since Sandra is using -ml-, Sandra may find a post I made a week or so 
ago about using -mfx- after user-written commands helpful:

http://www.stata.com/statalist/archive/2005-10/msg00113.html

 --May
mmb@stata.com
*
*   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/