Cross-Validatory Choice of the Number of Components From a Principal Component Analysis

A method is described for choosing the number of components to retain in a principal component analysis when the aim is dimensionality reduction. The correspondence between principal component analysis and the singular value decomposition of the data matrix is used. The method is based on successive...

Full description

Saved in:
Bibliographic Details
Published in:Technometrics 1982-02, Vol.24 (1), p.73-77
Main Authors: Eastment, H. T., Krzanowski, W. J.
Format: Article
Language:English
Subjects:
Coordinate systems
Correlations
Covariance matrices
Cross-validation
Datasets
Dimensionality reduction
Eigenvalues
Eigenvectors
Matrices
Modeling
Principal component analysis
Principal components analysis
Reduction of dimensionality
Singular value decomposition
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c321t-75938029a96bf27f8ed1d843f8a57be8607cfd4db762af0a4622c881de57ad3f3
cites cdi_FETCH-LOGICAL-c321t-75938029a96bf27f8ed1d843f8a57be8607cfd4db762af0a4622c881de57ad3f3
container_end_page 77
container_issue 1
container_start_page 73
container_title Technometrics
container_volume 24
creator Eastment, H. T.
Krzanowski, W. J.
description A method is described for choosing the number of components to retain in a principal component analysis when the aim is dimensionality reduction. The correspondence between principal component analysis and the singular value decomposition of the data matrix is used. The method is based on successively predicting each element in the data matrix after deleting the corresponding row and column of the matrix, and makes use of recently published algorithms for updating a singular value decomposition. These are very fast, which renders the proposed technique a practicable one for routine data analysis.
doi_str_mv 10.1080/00401706.1982.10487712
format article
fullrecord <record><control><sourceid>jstor_cross</sourceid><recordid>TN_cdi_crossref_primary_10_1080_00401706_1982_10487712</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><jstor_id>1267581</jstor_id><sourcerecordid>1267581</sourcerecordid><originalsourceid>FETCH-LOGICAL-c321t-75938029a96bf27f8ed1d843f8a57be8607cfd4db762af0a4622c881de57ad3f3</originalsourceid><addsrcrecordid>eNqFkE1LxDAQhoMouK7-BcnBa9d8tEl6XIqrwqIe_DiGaZOwXdpmSSrSf29LXfTmaZiZ553hfRG6pmRFiSK3hKSESiJWNFdsHKVKSspO0IJmXCZMMn6KFhOUTNQ5uohxTwjlTMkF-iiCjzF5h6Y20Psw4GLn68pi73C_s_jpsy1tmLrCtwff2a6PeBN8iwG_hLqr6gM0vzu87qAZYh0v0ZmDJtqrn7pEb5u71-Ih2T7fPxbrbVJxRvtEZjlXhOWQi9Ix6ZQ11KiUOwWZLK0SRFbOpKaUgoEjkArGKqWosZkEwx1fIjHfrSYfwTp9CHULYdCU6CkefYxHT_HoYzyj8GYW7uNo-6-KcSI1ZUJmio7YesbqzvnQwpcPjdE9DI0PLsDoP2r-z6tvdsd4OA</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Cross-Validatory Choice of the Number of Components From a Principal Component Analysis</title><source>JSTOR Archival Journals and Primary Sources Collection</source><creator>Eastment, H. T. ; Krzanowski, W. J.</creator><creatorcontrib>Eastment, H. T. ; Krzanowski, W. J.</creatorcontrib><description>A method is described for choosing the number of components to retain in a principal component analysis when the aim is dimensionality reduction. The correspondence between principal component analysis and the singular value decomposition of the data matrix is used. The method is based on successively predicting each element in the data matrix after deleting the corresponding row and column of the matrix, and makes use of recently published algorithms for updating a singular value decomposition. These are very fast, which renders the proposed technique a practicable one for routine data analysis.</description><identifier>ISSN: 0040-1706</identifier><identifier>EISSN: 1537-2723</identifier><identifier>DOI: 10.1080/00401706.1982.10487712</identifier><language>eng</language><publisher>Taylor &amp; Francis Group</publisher><subject>Coordinate systems ; Correlations ; Covariance matrices ; Cross-validation ; Datasets ; Dimensionality reduction ; Eigenvalues ; Eigenvectors ; Matrices ; Modeling ; Principal component analysis ; Principal components analysis ; Reduction of dimensionality ; Singular value decomposition</subject><ispartof>Technometrics, 1982-02, Vol.24 (1), p.73-77</ispartof><rights>Copyright Taylor &amp; Francis Group, LLC 1982</rights><rights>Copyright 1982 The American Statistical Association and the American Society for Quality Control</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c321t-75938029a96bf27f8ed1d843f8a57be8607cfd4db762af0a4622c881de57ad3f3</citedby><cites>FETCH-LOGICAL-c321t-75938029a96bf27f8ed1d843f8a57be8607cfd4db762af0a4622c881de57ad3f3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.jstor.org/stable/pdf/1267581$$EPDF$$P50$$Gjstor$$H</linktopdf><linktohtml>$$Uhttps://www.jstor.org/stable/1267581$$EHTML$$P50$$Gjstor$$H</linktohtml><link.rule.ids>314,780,784,27924,27925,58238,58471</link.rule.ids></links><search><creatorcontrib>Eastment, H. T.</creatorcontrib><creatorcontrib>Krzanowski, W. J.</creatorcontrib><title>Cross-Validatory Choice of the Number of Components From a Principal Component Analysis</title><title>Technometrics</title><description>A method is described for choosing the number of components to retain in a principal component analysis when the aim is dimensionality reduction. The correspondence between principal component analysis and the singular value decomposition of the data matrix is used. The method is based on successively predicting each element in the data matrix after deleting the corresponding row and column of the matrix, and makes use of recently published algorithms for updating a singular value decomposition. These are very fast, which renders the proposed technique a practicable one for routine data analysis.</description><subject>Coordinate systems</subject><subject>Correlations</subject><subject>Covariance matrices</subject><subject>Cross-validation</subject><subject>Datasets</subject><subject>Dimensionality reduction</subject><subject>Eigenvalues</subject><subject>Eigenvectors</subject><subject>Matrices</subject><subject>Modeling</subject><subject>Principal component analysis</subject><subject>Principal components analysis</subject><subject>Reduction of dimensionality</subject><subject>Singular value decomposition</subject><issn>0040-1706</issn><issn>1537-2723</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>1982</creationdate><recordtype>article</recordtype><recordid>eNqFkE1LxDAQhoMouK7-BcnBa9d8tEl6XIqrwqIe_DiGaZOwXdpmSSrSf29LXfTmaZiZ553hfRG6pmRFiSK3hKSESiJWNFdsHKVKSspO0IJmXCZMMn6KFhOUTNQ5uohxTwjlTMkF-iiCjzF5h6Y20Psw4GLn68pi73C_s_jpsy1tmLrCtwff2a6PeBN8iwG_hLqr6gM0vzu87qAZYh0v0ZmDJtqrn7pEb5u71-Ih2T7fPxbrbVJxRvtEZjlXhOWQi9Ix6ZQ11KiUOwWZLK0SRFbOpKaUgoEjkArGKqWosZkEwx1fIjHfrSYfwTp9CHULYdCU6CkefYxHT_HoYzyj8GYW7uNo-6-KcSI1ZUJmio7YesbqzvnQwpcPjdE9DI0PLsDoP2r-z6tvdsd4OA</recordid><startdate>19820201</startdate><enddate>19820201</enddate><creator>Eastment, H. T.</creator><creator>Krzanowski, W. J.</creator><general>Taylor &amp; Francis Group</general><general>The American Society for Quality Control and The American Statistical Association</general><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>19820201</creationdate><title>Cross-Validatory Choice of the Number of Components From a Principal Component Analysis</title><author>Eastment, H. T. ; Krzanowski, W. J.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c321t-75938029a96bf27f8ed1d843f8a57be8607cfd4db762af0a4622c881de57ad3f3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>1982</creationdate><topic>Coordinate systems</topic><topic>Correlations</topic><topic>Covariance matrices</topic><topic>Cross-validation</topic><topic>Datasets</topic><topic>Dimensionality reduction</topic><topic>Eigenvalues</topic><topic>Eigenvectors</topic><topic>Matrices</topic><topic>Modeling</topic><topic>Principal component analysis</topic><topic>Principal components analysis</topic><topic>Reduction of dimensionality</topic><topic>Singular value decomposition</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Eastment, H. T.</creatorcontrib><creatorcontrib>Krzanowski, W. J.</creatorcontrib><collection>CrossRef</collection><jtitle>Technometrics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Eastment, H. T.</au><au>Krzanowski, W. J.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Cross-Validatory Choice of the Number of Components From a Principal Component Analysis</atitle><jtitle>Technometrics</jtitle><date>1982-02-01</date><risdate>1982</risdate><volume>24</volume><issue>1</issue><spage>73</spage><epage>77</epage><pages>73-77</pages><issn>0040-1706</issn><eissn>1537-2723</eissn><abstract>A method is described for choosing the number of components to retain in a principal component analysis when the aim is dimensionality reduction. The correspondence between principal component analysis and the singular value decomposition of the data matrix is used. The method is based on successively predicting each element in the data matrix after deleting the corresponding row and column of the matrix, and makes use of recently published algorithms for updating a singular value decomposition. These are very fast, which renders the proposed technique a practicable one for routine data analysis.</abstract><pub>Taylor &amp; Francis Group</pub><doi>10.1080/00401706.1982.10487712</doi><tpages>5</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0040-1706
ispartof Technometrics, 1982-02, Vol.24 (1), p.73-77
issn 0040-1706
1537-2723
language eng
recordid cdi_crossref_primary_10_1080_00401706_1982_10487712
source JSTOR Archival Journals and Primary Sources Collection
subjects Coordinate systems
Correlations
Covariance matrices
Cross-validation
Datasets
Dimensionality reduction
Eigenvalues
Eigenvectors
Matrices
Modeling
Principal component analysis
Principal components analysis
Reduction of dimensionality
Singular value decomposition
title Cross-Validatory Choice of the Number of Components From a Principal Component Analysis
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-28T23%3A43%3A41IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-jstor_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Cross-Validatory%20Choice%20of%20the%20Number%20of%20Components%20From%20a%20Principal%20Component%20Analysis&rft.jtitle=Technometrics&rft.au=Eastment,%20H.%20T.&rft.date=1982-02-01&rft.volume=24&rft.issue=1&rft.spage=73&rft.epage=77&rft.pages=73-77&rft.issn=0040-1706&rft.eissn=1537-2723&rft_id=info:doi/10.1080/00401706.1982.10487712&rft_dat=%3Cjstor_cross%3E1267581%3C/jstor_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c321t-75938029a96bf27f8ed1d843f8a57be8607cfd4db762af0a4622c881de57ad3f3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_jstor_id=1267581&rfr_iscdi=true