I'm now studying for a partial least squares regression (PLS-R).
Unlike PCA, it says that score vectors are orthogonal, while loadings are not. Does this mean that loading vectors (which are actually just regression parameters) can happen to be linearly dependent?
At first I thought loading matrix should be linearly independent since if not, what's the point of the corresponding scores? Then after studying more I thought it may not be the case just like scores of PCA are generally not linearly independent.
However, I couldn't find a case through some experiment where loading vectors are linearly dependent (so that loading matrix is not full column rank).
Does the methodology of PLS guarantees its loading matrix to be linearly independent?
