Available in: GLM, GAM, CoxPH
By default, interactions between predictor columns are expanded and computed on the fly as GLM iterates over dataset. The
interactions option allows you to enter a list of predictor column indices that should interact.
Note that adding a list of interactions to a model changes the interpretation of all of the coefficients. For example, a typical predictor has the form ‘response ~ terms’ where ‘response’ is the (numeric) response vector, and ‘terms’ is a series of terms that specify a linear predictor for ‘response’. For ‘binomial’ and ‘quasibinomial’ families in GLM, the response can also be specified as a ‘factor’ (when the first level denotes failure and all other levels denote success) or as a two-column matrix with the columns giving the numbers of successes and failures.
Interactions can be specified between two categorical columns, between two numeric columns, or between a mix of categorical and numerical columns. When entered, all pairwise combinations of predictor column indices will be computed for that list.