[R] Quadratic function with interaction terms for the PLS fitting model?

[Bert Gunter]

> It would seem reasonable that the help for poly() could make it explicitly clear that if 'x' is not a vector, but is a matrix, that 'degree' must be explicitly named.
>
> Regards,
>
> Marc
>

Exactly.  As written, there is no reason to believe that the stated
exception to the ... argument matching rule does not also apply to x a
matrix.

-- Bert