Dear Statalister,
I am estimating a regoprob and I would like to compute the average
partial effects.
To do this I need to rescale the parameters before I obtain the
predicted probabilities.
By having a look at the regoprob_p.ado (line 284 onwards), when the
-predict p- command is used, the linear prediction is rescaled
automatically and then the predicted probabilities are computed.
However, I think that when the option -predict xb- is used, the linear
prediction is not rescaled and in order to move on I have to rescale it
myself. Is that correct?
Finally, I d like to ask how exactly the linear prediction is computed
from regorpob if one wanted to do it manually. When I am trying, I get a
constant difference between my computation and the one from regoprob
that, I think, shows that I am missing something very specific.
Thank you in advance
Regards
Manos
*
* 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/