I have a question about the use of the Stata marginal effects command
'mfx' and the module 'ivprob' written by Joseph Harkness. I am using
Stata version 7.
1) Use of the predict command after 'ivprob' generates linear
predictions. That is, 'predict' and 'predict ,xb' generate the same
result. (This is not a problem since you can get probit predictions
by applying the 'normprob' function to the linear prediction.)
2) Use of the 'mfx compute' and 'mfx compute, predict(xb)' commands
do NOT generate the same results. Why? The Stata output says both
are linear predictions. Does 'mfx compute' produce meaningful
results at all with 'ivprob'? How should I interpret these?
Thanks,
Derek Brown
See http://ideas.repec.org/c/boc/bocode/s415801.html for more on 'ivprob'.
*
* 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/