A logit is the natural logarithm of an event’s odds: log(p/(1 − p)), where p is the event probability and 1 − p is its complement. Logistic regression models the log odds of a binary outcome rather than modeling the outcome directly. section 5.0 section 5.1
In logistic regression, each coefficient estimates the change in log odds associated with a one-unit increase in its predictor. The model then passes the resulting linear combination through the sigmoid function to obtain a probability between 0 and 1 for an outcome coded 0 or 1. section 5.1
Interpreting a logistic-regression coefficient
For a predictor X with coefficient β, increasing X by one unit changes the modeled log odds by β. The sigmoid transformation converts the model’s resulting value into the predicted probability of the binary outcome. section 5.1
