In the previous section, we noted that linear regression is a good choice when the target variable is continuous. We're now going to move on to look at a binomial logistic regression model, which can predict the probability that an observation falls into one of two categories of a dichotomous target variable based on one or more predictor variables. A binomial logistic regression is often referred to as logistic regression.
Logistic regression is similar to linear regression, except that the dependent variable is measured on a dichotomous scale. Logistic regression allows us to model a relationship between multiple predictor variables and a dichotomous target variable. However, unlike linear regression, in the case of logistic regression, the linear function is used as an input to another function, such as 
:
Here, is the sigmoid or logistic function...
