Re:Logistic regression from scratch
by delib on Thursday June 25, 2020 @12:57PM
(#23642751)
> it's linear regression on the log-odds
Almost - logistic regression assumes that the function is linear in the log odds, i.e. log(p/(1-p)) = Xb + e. The problem is that you can't compute the log-odds, because you don't know p.
Almost - logistic regression assumes that the function is linear in the log odds, i.e. log(p/(1-p)) = Xb + e. The problem is that you can't compute the log-odds, because you don't know p.