Sunday, February 26, 2006
Thanks to Kit Baum, a new version of -pre- has been released.
-pre- is a post-estimation command that computes the Proportional Reduction of Errors, a measure of a models predictive power. It is simply R-squared for OLS regression.
The new version of -pre- will, after logit/probit models, compute the value for the cutoff (usually set at 0.5) that maximizes the number of correct outcomes. This new version is also more flexible, allowing the computation of the Proportion Reduction of Errors based on the user's choice of central tendency. It also fixes a few bugs. The new version requires Stata 9.
To install type: ssc install pre, replace
An example: sysuse auto, clear logit foreign rep78 price turn pre pre, maxcut
* * 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/
Links to this post: