QR-PLSR - REDUCED-RANK REGRESSION FOR HIGH-SPEED HARDWARE IMPLEMENTATION

Authors
WESTAD F DIEPOLD K MARTENS H
Citation
F. Westad et al., QR-PLSR - REDUCED-RANK REGRESSION FOR HIGH-SPEED HARDWARE IMPLEMENTATION, Journal of chemometrics, 10(5-6), 1996, pp. 439-451
Citations number
19
Categorie Soggetti
Chemistry Analytical","Statistic & Probability
Journal title
Journal of chemometrics → ACNP
ISSN journal
08869383
Volume
10
Issue
5-6
Year of publication
1996
Pages
439 - 451
Database
ISI
SICI code
0886-9383(1996)10:5-6<439:Q-RRFH>2.0.ZU;2-6
Abstract
A version of PLS regression is described that intends to combine the c omputer hardware implementation advantages of the algebraic technique of 'QR decomposition' with the statistical, interpretative and computa tional advantages of PLS regression. With a QR decomposition based on Givens rotations, the QR-PLS technique appears to be suited for hardwa re parallelization without sacrificing the modelling flexibility of PL SR. (C) 1996 by John Wiley & Sons, Ltd.