st: Calculating Standard Errors of Predicted Probabilities in ProbitModels

[Martin Daniel Heintzelman <mheintze@umich.edu>]

I am running a probit model with towns in Massachusetts as my 
observations, and the dependent variable being whether or not the town as 
implemented a particular policy.  After running the model I calculate the 
predicted probabilities for all towns.  I want to compare these 
probabilities for TWO specific towns  - are they significantly different 
or not?  I was hoping to perform a t-test to figure this out, but need the 
standard errors of the predicted probabilities to do this.

The predict command will give me the standard errors of the linear 
prediction - is there a way, from that information, to calculate the 
standard error of the predicted probability?  Or is there some OTHER way 
to test the hypothesis that these predicted probabilities are the same (or 
different)?

Thanks for your help!

Martin Heintzelman
Ph.D. Candidate
Department of Economics
University of Michigan
*
*   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/