ABSTRACT
This chapter builds in the terminology and concepts introduced in the linear regression chapter. In binary logistic regression the outcome must have two levels, but like in linear regression, the predictors (explanatory variables) can be either continuous or categorical. The chapter covers some further concepts such as odds vs probabilities, odds ratios, logit, and revisits the terms shared with linear regression, such as interactions, additive, multiplicative. The chapter goes into more depth around some of the assumptions, especially linearity and collinearity, and introduces further model metrics.
