Multivariate seeded dimension reduction

Jae Keun Yoo; Yunju Im

doi:10.1016/j.jkss.2014.03.002

Multivariate seeded dimension reduction

Jae Keun Yoo, Yunju Im

Department of Statistics

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

5 Scopus citations

Abstract

A recently introduced seeded dimension reduction approach enables existing sufficient dimension reduction methods to be used in regressions with n<. p. The dimension reduction is accomplished through successive projections of seed matrices on a subspace to contain the central subspace. In the article, we will develop a seeded dimension reduction for multivariate regression, whose responses are multi-dimensional. For this we suggest two conditions that the dimension reduction is attained without the loss of information of the central subspace. Based on this, we construct possible candidate seed matrices. Numerical studies and two data analyses are presented.

Original language	English
Pages (from-to)	559-566
Number of pages	8
Journal	Journal of the Korean Statistical Society
Volume	43
Issue number	4
DOIs	https://doi.org/10.1016/j.jkss.2014.03.002
State	Published - 1 Dec 2014

Bibliographical note

Funding Information:
For Yunju Im, this work was supported by the BK21 Plus Project through the National Research Foundation of Korea (NRF) funded by the Korean Ministry of Education ( 22A20130011003 ).

Funding Information:
For Jae Keun Yoo, this works was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Korean Ministry of Education ( NRF-2012R1A1A1040077 ).

Publisher Copyright:
© 2014 The Korean Statistical Society.

Keywords

Large p small n
Multivariate regression
Seed matrix
Sufficient dimension reduction

Access to Document

10.1016/j.jkss.2014.03.002

Cite this

@article{8445f961d2bd407dabfe5f35a7dc6c2b,

title = "Multivariate seeded dimension reduction",

abstract = "A recently introduced seeded dimension reduction approach enables existing sufficient dimension reduction methods to be used in regressions with n<. p. The dimension reduction is accomplished through successive projections of seed matrices on a subspace to contain the central subspace. In the article, we will develop a seeded dimension reduction for multivariate regression, whose responses are multi-dimensional. For this we suggest two conditions that the dimension reduction is attained without the loss of information of the central subspace. Based on this, we construct possible candidate seed matrices. Numerical studies and two data analyses are presented.",

keywords = "Large p small n, Multivariate regression, Seed matrix, Sufficient dimension reduction",

author = "Yoo, {Jae Keun} and Yunju Im",

note = "Funding Information: For Yunju Im, this work was supported by the BK21 Plus Project through the National Research Foundation of Korea (NRF) funded by the Korean Ministry of Education ( 22A20130011003 ). Funding Information: For Jae Keun Yoo, this works was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Korean Ministry of Education ( NRF-2012R1A1A1040077 ). Publisher Copyright: {\textcopyright} 2014 The Korean Statistical Society.",

year = "2014",

month = dec,

day = "1",

doi = "10.1016/j.jkss.2014.03.002",

language = "English",

volume = "43",

pages = "559--566",

journal = "Journal of the Korean Statistical Society",

issn = "1226-3192",

publisher = "Korean Statistical Society",

number = "4",

}

TY - JOUR

T1 - Multivariate seeded dimension reduction

AU - Yoo, Jae Keun

AU - Im, Yunju

N1 - Funding Information: For Yunju Im, this work was supported by the BK21 Plus Project through the National Research Foundation of Korea (NRF) funded by the Korean Ministry of Education ( 22A20130011003 ). Funding Information: For Jae Keun Yoo, this works was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Korean Ministry of Education ( NRF-2012R1A1A1040077 ). Publisher Copyright: © 2014 The Korean Statistical Society.

PY - 2014/12/1

Y1 - 2014/12/1

N2 - A recently introduced seeded dimension reduction approach enables existing sufficient dimension reduction methods to be used in regressions with n<. p. The dimension reduction is accomplished through successive projections of seed matrices on a subspace to contain the central subspace. In the article, we will develop a seeded dimension reduction for multivariate regression, whose responses are multi-dimensional. For this we suggest two conditions that the dimension reduction is attained without the loss of information of the central subspace. Based on this, we construct possible candidate seed matrices. Numerical studies and two data analyses are presented.

AB - A recently introduced seeded dimension reduction approach enables existing sufficient dimension reduction methods to be used in regressions with n<. p. The dimension reduction is accomplished through successive projections of seed matrices on a subspace to contain the central subspace. In the article, we will develop a seeded dimension reduction for multivariate regression, whose responses are multi-dimensional. For this we suggest two conditions that the dimension reduction is attained without the loss of information of the central subspace. Based on this, we construct possible candidate seed matrices. Numerical studies and two data analyses are presented.

KW - Large p small n

KW - Multivariate regression

KW - Seed matrix

KW - Sufficient dimension reduction

UR - http://www.scopus.com/inward/record.url?scp=84908501185&partnerID=8YFLogxK

U2 - 10.1016/j.jkss.2014.03.002

DO - 10.1016/j.jkss.2014.03.002

M3 - Article

AN - SCOPUS:84908501185

SN - 1226-3192

VL - 43

SP - 559

EP - 566

JO - Journal of the Korean Statistical Society

JF - Journal of the Korean Statistical Society

IS - 4

ER -

Multivariate seeded dimension reduction

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this