Hi R community,
I am trying to fit a PLS model with response Y and predictor X, where X consists
of at least 30 columns (say x1, x2, ..., x30). Aside from studying the
relationship of Y and x1, ..., x30, I am also interested in studying the effect
of quadratic terms (x1^2, ..., x30^2) and two-way interactions (x1*x2, x1*x3,
..., x29*x30) on Y.
My R code is
dat.pls<-plsr(Y ~ X^2 + I(X^2), ncomp = 2, method="oscorespls")
This code does not work for me. I got this from Chapter 11 of "An
Introduction to R."
Any suggestion of what the appropritae code should be is highly appreciated.
Thanks,
Jaclyn M.
[[alternative HTML version deleted]]
