A robust implementation of partial least squares (PLS) is developed in
which the method of iteratively reweighted least squares is adapted f
or use with PLS. The result is a PLS algorithm which is robust to outl
iers and is easy to implement. Examples and case studies are presented
, followed by two Monte Carlo studies designed to explore the behavior
of the method. The paper begins with the motivation and intended appl
ications for the procedure. A discussion is given of the method of ite
ratively reweighted least squares (IRLS) for outlier detection. The pr
ocedure, given the name IRPLS, is then presented. Three case studies i
llustrate how the procedure works on various types of data and how it
should be used. The first Monte Carlo study is designed to determine w
hether the IRPLS procedure correctly identifies multiple outliers in a
wide variety of configurations. The second Monte Carlo study is desig
ned to estimate the breakdown bound of the procedure.