In this model, we have made no assumptions about the distributions of the predictor variables, the Xi. They can be continuously measured, they can be contrast-coded predictors used to represent categorical variables, or they can be products of variables. We have, however, made assumptions all along about the residuals, εi, in this model. Speciﬁcally, we have assumed three things about εi throughout the preceding chapters:
1. They are normally distributed. 2. They have constant variance. 3. They are independent of each other.