Quantcast
Channel: The Stata Blog - Latest Comments
Viewing all articles
Browse latest Browse all 1126

Re: probit or logit: ladies and gentlemen, pick your weapon

$
0
0

I've preferred logistic regressions ever since I learned that maximum size of marginal effects in a logistic regression is beta/4, so it is easy to know if the marginal effects could have real world significance without a calculator. You can do the same thing with a probit but the maximum effect is beta/(1.7 * 4), which I can't do nearly as easily.


Viewing all articles
Browse latest Browse all 1126

Trending Articles