We give a geometric proof that the estimates of a regression model derived by using partial least squares shrink the ordinary least squares estimates. The proof is based on a sequential construction algorithm of partial least squares. A discussion of the nature of shrinkage is included.
Publié le : 1996-04-14
Classification:
Biased estimation,
chemometrics,
collinearity,
data compression methods,
ordinary least squares,
regression,
62J07,
62F10
@article{1032894467,
author = {Goutis, Constantinos},
title = {Partial least squares algorithm yields shrinkage estimators},
journal = {Ann. Statist.},
volume = {24},
number = {6},
year = {1996},
pages = { 816-824},
language = {en},
url = {http://dml.mathdoc.fr/item/1032894467}
}
Goutis, Constantinos. Partial least squares algorithm yields shrinkage estimators. Ann. Statist., Tome 24 (1996) no. 6, pp. 816-824. http://gdmltest.u-ga.fr/item/1032894467/