Optimal sufficient dimension reduction for the conditional mean in multivariate regression

Jae Keun Yoo; R. Dennis Cook

doi:10.1093/biomet/asm003

Optimal sufficient dimension reduction for the conditional mean in multivariate regression

Jae Keun Yoo, R. Dennis Cook

Department of Statistics

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

26 Scopus citations

Abstract

The aim of this article is to develop optimal sufficient dimension reduction methodology for the conditional mean in multivariate regression. The context is roughly the same as that of a related method by Cook & Setodji (2003), but the new method has several advantages. It is asymptotically optimal in the sense described herein and its test statistic for dimension always has a chi-squared distribution asymptotically under the null hypothesis. Additionally, the optimal method allows tests of predictor effects. A comparison of the two methods is provided.

Original language	English
Pages (from-to)	231-242
Number of pages	12
Journal	Biometrika
Volume	94
Issue number	1
DOIs	https://doi.org/10.1093/biomet/asm003
State	Published - Mar 2007

Bibliographical note

Funding Information:
ACKNOWLEDGEMENT The authors are grateful to the referees for many helpful comments. This work was supported in part by grants from the U.S. National Science Foundation.

Keywords

Multivariate conditional mean
Multivariate regression
Predictor effect test
Sufficient dimension reduction

Access to Document

10.1093/biomet/asm003

Cite this

@article{648a4f3ed92a41d0919617017c26b0f3,

title = "Optimal sufficient dimension reduction for the conditional mean in multivariate regression",

abstract = "The aim of this article is to develop optimal sufficient dimension reduction methodology for the conditional mean in multivariate regression. The context is roughly the same as that of a related method by Cook & Setodji (2003), but the new method has several advantages. It is asymptotically optimal in the sense described herein and its test statistic for dimension always has a chi-squared distribution asymptotically under the null hypothesis. Additionally, the optimal method allows tests of predictor effects. A comparison of the two methods is provided.",

keywords = "Multivariate conditional mean, Multivariate regression, Predictor effect test, Sufficient dimension reduction",

author = "Yoo, {Jae Keun} and Cook, {R. Dennis}",

note = "Funding Information: ACKNOWLEDGEMENT The authors are grateful to the referees for many helpful comments. This work was supported in part by grants from the U.S. National Science Foundation.",

year = "2007",

month = mar,

doi = "10.1093/biomet/asm003",

language = "English",

volume = "94",

pages = "231--242",

journal = "Biometrika",

issn = "0006-3444",

publisher = "Oxford University Press",

number = "1",

}

TY - JOUR

T1 - Optimal sufficient dimension reduction for the conditional mean in multivariate regression

AU - Yoo, Jae Keun

AU - Cook, R. Dennis

N1 - Funding Information: ACKNOWLEDGEMENT The authors are grateful to the referees for many helpful comments. This work was supported in part by grants from the U.S. National Science Foundation.

PY - 2007/3

Y1 - 2007/3

N2 - The aim of this article is to develop optimal sufficient dimension reduction methodology for the conditional mean in multivariate regression. The context is roughly the same as that of a related method by Cook & Setodji (2003), but the new method has several advantages. It is asymptotically optimal in the sense described herein and its test statistic for dimension always has a chi-squared distribution asymptotically under the null hypothesis. Additionally, the optimal method allows tests of predictor effects. A comparison of the two methods is provided.

AB - The aim of this article is to develop optimal sufficient dimension reduction methodology for the conditional mean in multivariate regression. The context is roughly the same as that of a related method by Cook & Setodji (2003), but the new method has several advantages. It is asymptotically optimal in the sense described herein and its test statistic for dimension always has a chi-squared distribution asymptotically under the null hypothesis. Additionally, the optimal method allows tests of predictor effects. A comparison of the two methods is provided.

KW - Multivariate conditional mean

KW - Multivariate regression

KW - Predictor effect test

KW - Sufficient dimension reduction

UR - http://www.scopus.com/inward/record.url?scp=33947163989&partnerID=8YFLogxK

U2 - 10.1093/biomet/asm003

DO - 10.1093/biomet/asm003

M3 - Article

AN - SCOPUS:33947163989

SN - 0006-3444

VL - 94

SP - 231

EP - 242

JO - Biometrika

JF - Biometrika

IS - 1

ER -

Optimal sufficient dimension reduction for the conditional mean in multivariate regression

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this