A new reduced-rank linear discriminant analysis method and its applications

Yue Selena Niu; Ning Hao; Bin Dong

doi:10.5705/ss.202015.0387

A new reduced-rank linear discriminant analysis method and its applications

Yue Selena Niu, Ning Hao, Bin Dong

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

13 Scopus citations

Abstract

We consider multi-class classification problems for high-dimensional data. Following the idea of reduced-rank linear discriminant analysis (LDA), we introduce a new dimension reduction tool with a flavor of supervised principal component analysis (PCA). The proposed method is computationally efficient and can incorporate the correlation structure among the features. Besides the theoretical insights, we show that our method is a competitive classification tool by simulated and real data examples.

Original language	English (US)
Pages (from-to)	189-202
Number of pages	14
Journal	Statistica Sinica
Volume	28
Issue number	1
DOIs	https://doi.org/10.5705/ss.202015.0387
State	Published - Jan 2018
Externally published	Yes

Keywords

Dimension reduction
Gene expression data
High-dimensional data
Multi-class classification
Supervised principal component analysis

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty

Access to Document

10.5705/ss.202015.0387

Cite this

@article{1867ea4a5ff745aea1ce78ab9ff2a02a,

title = "A new reduced-rank linear discriminant analysis method and its applications",

abstract = "We consider multi-class classification problems for high-dimensional data. Following the idea of reduced-rank linear discriminant analysis (LDA), we introduce a new dimension reduction tool with a flavor of supervised principal component analysis (PCA). The proposed method is computationally efficient and can incorporate the correlation structure among the features. Besides the theoretical insights, we show that our method is a competitive classification tool by simulated and real data examples.",

keywords = "Dimension reduction, Gene expression data, High-dimensional data, Multi-class classification, Supervised principal component analysis",

author = "Niu, {Yue Selena} and Ning Hao and Bin Dong",

note = "Funding Information: This research was partially supported by grants DMS 1309507 from the National Science Foundation and 11671022 from the National Science Foundation of China. The authors are grateful to Daniela Witten for helpful discussions, and to the Editor, associate editor, and two referees for comments and suggestions that greatly improved the paper.",

year = "2018",

month = jan,

doi = "10.5705/ss.202015.0387",

language = "English (US)",

volume = "28",

pages = "189--202",

journal = "Statistica Sinica",

issn = "1017-0405",

publisher = "Institute of Statistical Science",

number = "1",

}

TY - JOUR

T1 - A new reduced-rank linear discriminant analysis method and its applications

AU - Niu, Yue Selena

AU - Hao, Ning

AU - Dong, Bin

N1 - Funding Information: This research was partially supported by grants DMS 1309507 from the National Science Foundation and 11671022 from the National Science Foundation of China. The authors are grateful to Daniela Witten for helpful discussions, and to the Editor, associate editor, and two referees for comments and suggestions that greatly improved the paper.

PY - 2018/1

Y1 - 2018/1

N2 - We consider multi-class classification problems for high-dimensional data. Following the idea of reduced-rank linear discriminant analysis (LDA), we introduce a new dimension reduction tool with a flavor of supervised principal component analysis (PCA). The proposed method is computationally efficient and can incorporate the correlation structure among the features. Besides the theoretical insights, we show that our method is a competitive classification tool by simulated and real data examples.

AB - We consider multi-class classification problems for high-dimensional data. Following the idea of reduced-rank linear discriminant analysis (LDA), we introduce a new dimension reduction tool with a flavor of supervised principal component analysis (PCA). The proposed method is computationally efficient and can incorporate the correlation structure among the features. Besides the theoretical insights, we show that our method is a competitive classification tool by simulated and real data examples.

KW - Dimension reduction

KW - Gene expression data

KW - High-dimensional data

KW - Multi-class classification

KW - Supervised principal component analysis

UR - http://www.scopus.com/inward/record.url?scp=85040132111&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85040132111&partnerID=8YFLogxK

U2 - 10.5705/ss.202015.0387

DO - 10.5705/ss.202015.0387

M3 - Article

AN - SCOPUS:85040132111

SN - 1017-0405

VL - 28

SP - 189

EP - 202

JO - Statistica Sinica

JF - Statistica Sinica

IS - 1

ER -

A new reduced-rank linear discriminant analysis method and its applications

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this