PLS SHRINKS

被引：38

作者：

DEJONG, S

机构：

[1] Unilever Research Laboratorium Vlaardingen, Vlaardingen, 3130 AC

来源：

JOURNAL OF CHEMOMETRICS | 1995年 / 9卷 / 04期

关键词：

PARTIAL LEAST SQUARES; BIASED REGRESSION; ORDINARY LEAST SQUARES; MINIMUM LENGTH LEAST SQUARES;

D O I：

10.1002/cem.1180090406

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

An algebraic proof is given that in partial least squares (PLS) regression the Euclidean length of the estimator is shrunk in comparison with the ordinary least squares estimator or with PLS estimators based on a larger number of dimensions.

引用

页码：323 / 326

页数：4