REG

REG(f;s;m) indicates that the effects of factor f are to be partitioned into the orthogonal regression contrasts specified by the first s rows of the matrix m. In regression models, the first argument may be a variate instead of a factor; REG(v;s;m) then orthogonalizes and fits a set of associated variates stored in the first s rows of the rows of the matrix m. The matrix m may be omitted in a regression model, in which case orthogonal polynomial contrasts are constructed for either f or v. Note, though, that the orthogonalization is with respect to the replication of the main effect of the factor or variate, so interactions of the contrasts with other vectors in a regression model may not be orthogonal.