Using observed confidence levels to perform principal component analyses

Alan M. Polansky, Santu Ghosh

Research output: Contribution to journalArticle

Abstract

ABSTRACT: This paper focuses on applying the method of observed confidence levels to problems commonly encountered in principal component analyses. In particular, we focus on assigning levels of confidence to the number of components that explain a specified proportion of variation in the original data. Approaches based on the normal model as well as a non parametric model are explored. The usefulness of the methods are discussed using an example and an empirical study.

Original languageEnglish (US)
Pages (from-to)3596-3611
Number of pages16
JournalCommunications in Statistics - Theory and Methods
Volume45
Issue number12
DOIs
StatePublished - Jun 17 2016

Fingerprint

Confidence Level
Principal Components
Nonparametric Model
Number of Components
Empirical Study
Confidence
Proportion
Model

Keywords

  • Eigenvalue
  • Eigenvector
  • Multivariate analysis
  • Non parametric bootstrap
  • Parametric bootstrap

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Using observed confidence levels to perform principal component analyses. / Polansky, Alan M.; Ghosh, Santu.

In: Communications in Statistics - Theory and Methods, Vol. 45, No. 12, 17.06.2016, p. 3596-3611.

Research output: Contribution to journalArticle

@article{2a3ef1ea17a644f084360515c37f47dd,
title = "Using observed confidence levels to perform principal component analyses",
abstract = "ABSTRACT: This paper focuses on applying the method of observed confidence levels to problems commonly encountered in principal component analyses. In particular, we focus on assigning levels of confidence to the number of components that explain a specified proportion of variation in the original data. Approaches based on the normal model as well as a non parametric model are explored. The usefulness of the methods are discussed using an example and an empirical study.",
keywords = "Eigenvalue, Eigenvector, Multivariate analysis, Non parametric bootstrap, Parametric bootstrap",
author = "Polansky, {Alan M.} and Santu Ghosh",
year = "2016",
month = "6",
day = "17",
doi = "10.1080/03610926.2014.904356",
language = "English (US)",
volume = "45",
pages = "3596--3611",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "12",

}

TY - JOUR

T1 - Using observed confidence levels to perform principal component analyses

AU - Polansky, Alan M.

AU - Ghosh, Santu

PY - 2016/6/17

Y1 - 2016/6/17

N2 - ABSTRACT: This paper focuses on applying the method of observed confidence levels to problems commonly encountered in principal component analyses. In particular, we focus on assigning levels of confidence to the number of components that explain a specified proportion of variation in the original data. Approaches based on the normal model as well as a non parametric model are explored. The usefulness of the methods are discussed using an example and an empirical study.

AB - ABSTRACT: This paper focuses on applying the method of observed confidence levels to problems commonly encountered in principal component analyses. In particular, we focus on assigning levels of confidence to the number of components that explain a specified proportion of variation in the original data. Approaches based on the normal model as well as a non parametric model are explored. The usefulness of the methods are discussed using an example and an empirical study.

KW - Eigenvalue

KW - Eigenvector

KW - Multivariate analysis

KW - Non parametric bootstrap

KW - Parametric bootstrap

UR - http://www.scopus.com/inward/record.url?scp=84969972548&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84969972548&partnerID=8YFLogxK

U2 - 10.1080/03610926.2014.904356

DO - 10.1080/03610926.2014.904356

M3 - Article

AN - SCOPUS:84969972548

VL - 45

SP - 3596

EP - 3611

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 12

ER -