Partial least squares for simultaneous reduction of response and predictor vectors in regression

R. Dennis Cook; Liliana Forzani; Lan Liu

doi:10.1016/j.jmva.2023.105163

Partial least squares for simultaneous reduction of response and predictor vectors in regression

R. Dennis Cook, Liliana Forzani, Lan Liu

Statistics (Twin Cities)

Research output: Contribution to journal › Article › peer-review

4 Scopus citations

Abstract

We study and establish a foundation for dimension reduction methods that compress the response and predictor vectors in multivariate regression. While all of the methods studied can perform competitively, depending on the characteristics of the regression, using partial least squares to compress the response and predictor vectors was judged to be the best for prediction and parameter estimation.

Original language	English (US)
Article number	105163
Journal	Journal of Multivariate Analysis
Volume	196
DOIs	https://doi.org/10.1016/j.jmva.2023.105163
State	Published - Jul 2023

Bibliographical note

Publisher Copyright:
© 2023

Keywords

Envelopes
NIPALS
Sufficient dimension reduction
Two-block method

Publisher link

10.1016/j.jmva.2023.105163

Find It

Find It at U of M Libraries

Cite this

@article{312f50da82354193b1e915a056c48b2f,

title = "Partial least squares for simultaneous reduction of response and predictor vectors in regression",

abstract = "We study and establish a foundation for dimension reduction methods that compress the response and predictor vectors in multivariate regression. While all of the methods studied can perform competitively, depending on the characteristics of the regression, using partial least squares to compress the response and predictor vectors was judged to be the best for prediction and parameter estimation.",

keywords = "Envelopes, NIPALS, Sufficient dimension reduction, Two-block method",

author = "Cook, {R. Dennis} and Liliana Forzani and Lan Liu",

note = "Publisher Copyright: {\textcopyright} 2023",

year = "2023",

month = jul,

doi = "10.1016/j.jmva.2023.105163",

language = "English (US)",

volume = "196",

journal = "Journal of Multivariate Analysis",

issn = "0047-259X",

publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Partial least squares for simultaneous reduction of response and predictor vectors in regression

AU - Cook, R. Dennis

AU - Forzani, Liliana

AU - Liu, Lan

PY - 2023/7

Y1 - 2023/7

N2 - We study and establish a foundation for dimension reduction methods that compress the response and predictor vectors in multivariate regression. While all of the methods studied can perform competitively, depending on the characteristics of the regression, using partial least squares to compress the response and predictor vectors was judged to be the best for prediction and parameter estimation.

AB - We study and establish a foundation for dimension reduction methods that compress the response and predictor vectors in multivariate regression. While all of the methods studied can perform competitively, depending on the characteristics of the regression, using partial least squares to compress the response and predictor vectors was judged to be the best for prediction and parameter estimation.

KW - Envelopes

KW - NIPALS

KW - Sufficient dimension reduction

KW - Two-block method

UR - http://www.scopus.com/inward/record.url?scp=85153395370&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85153395370&partnerID=8YFLogxK

U2 - 10.1016/j.jmva.2023.105163

DO - 10.1016/j.jmva.2023.105163

M3 - Article

AN - SCOPUS:85153395370

SN - 0047-259X

VL - 196

JO - Journal of Multivariate Analysis

JF - Journal of Multivariate Analysis

M1 - 105163

ER -

Partial least squares for simultaneous reduction of response and predictor vectors in regression

Abstract

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this