Dear List,
Following use of the Heckprob procedure, I have come across a challenge in
terms of calculating confidence intervals associated with predicted
probabilities. The procedure seems fairly straightforward for calculating
CIs around first stage predicted probabilities, however much less so at the
second stage.
What I am interested in is being able to produce estimates of predicted
probabilities and confidence intervals, based on heckprob results. My
ultimate goal is to be able to do this for various profiles of independent
variable values.
Any guidance would be appreciated.
Best,
Paul
Paul D. Senese, Ph. D.
Associate Professor
Department of Political Science
520 Park Hall
University at Buffalo, SUNY
Buffalo, NY 14260
716.645.2251, ext. 508 (phone)
716.645.2166 (fax) [email protected]
