Question

In a few sentences, explain to me why the two key conditions for fitting a logistic regression model are important to have.

Answer 1

The outcome variable of general logistic model is

where, i is the index used for the i-th observation.

The predictor variables are represented by

.

Now, Yitakes value 1 with probability pi; So, Yi is a two level categorical variable.

The two main conditions for fitting a logistic regression model are

(1)

Each predictor xi is linearly related to logit(pi) if all other predictors are held constant.

And,

(2)

Each outcome Yi is independent of the other outcomes.

So, the outcome variables are independent.