Question

I am doing a multiple regression in MATLAB. I am using regstats with the ‘interactions’ model type, and I have about 12 predictor variables. My understanding is that ‘interactions’ will perform the regression with all of the interaction terms. I assume that the order of the new regression variables is a constant, followed by all of the linear terms, followed by all of the interaction terms, but I don’t know what order the interaction terms are in.

I know that I can use x2fx with ‘interaction’ to generate the actual values, but then I have 66 interaction variables, and I can’t figure out which is which.

Is there any easy way to figure this out? Is there a way to generate the model that x2fx uses when you select ‘interactions’?

Answer 1

If X has n columns, the order of the columns of D for a full quadratic
model is:

1. The constant term
2. The linear terms (the columns of X, in order 1, 2, …, n)
3. The interaction terms (pairwise products of the columns of X, in order (1, 2), (1, 3), …, (1, n), (2, 3), …, (n–1, n))
4. The squared terms (in order 1, 2, …, n)

Source