Thanks to Kit Baum, a new version of -pre- has now been released.
This routine allows the tweaking of binary models by providing an optimum cut point for the probability that maximizes correct predictions.
Rare events
For modeling of rare events, the prediction of one is often more important than the prediction of a zero.
The new version allow for heavier weighting for prediction of one (compared to zero).
More efficient
The new version is considerably more efficient and allows for finding the cut point to fewer decimal places, which also decreases computer time (by bearly ten-fold per decimal place).
*
* 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/
