Logistic regression is a widely used statistical model that can be used to model a binary dependent variable. In linear regression, we assumed that the dependent variable is a numerical value that we were trying to predict. Consider a case where the binary variable has values of true and false. In logistic regression, instead of calculating the value of numerical output using the formula we used in the Linear regression section, we estimate the log odds of a binary event labeled True using the same formulation. The function that converts log odds to the probability of the event labeled 1 occurring is called the logistic function.
The unit of measurement for log-odds scale is called logit. Log-odds are calculated using the following formula:
Thus, using the same methodology as linear regression, logistic regression is used...
