A ROBUST PLS PROCEDURE

A robust partial least squares (PLS) regression algorithm is developed. This is achieved by substitution of the univariate regression steps in the iterative PLS2 algorithm by a robust alternative. The angle between loading vectors from both perturbed and unperturbed solutions is used as a measure of robustness. By means of a perturbation study on a structure-activity data set, it is demonstrated that the stability of the robust method is superior to standard PLS.