IMPROVED PLS ALGORITHMS

In this paper a proof is given that only one of either the X- or the Y -matrix in PLS algorithms needs to be deflated during the sequential p rocess of computing latent vectors. With the aid of this proof the ori ginal kernel algorithm developed by Lindgren et al. (J. Chemometrics, 7, 45 (1993)) is modified to provide two faster and more economical al gorithms. The performances of these new algorithms are compared with t hat of De Jong and Ter Braak's (J. Chemometrics, 8, 169 (1994)) modifi ed kernel algorithm in terms of speed and the new algorithms are shown to be much faster. A very fast kernel algorithm for updating PLS mode ls in a recursive manner and for exponentially discounting past data i s also presented. (C) 1997 by John Wiley & Sons, Ltd.