Logistic Regression
Logistic regression is an algorithm for linear and binary classification problems. It is a classification, not regression algorithm.
1. From odds ratio to sigmoidal logistic function If p is the probability of a sample being a positive event, then the odds ratio is defined as: odds ratio=p1−p.
The logarithm of the odds ratio function is called logit function: z=logit(p)=log(p1−p).
Logit function maps the probability {p|p∈[0,1]} to the entire real domain {z|z∈(−∞,+∞)}.