File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Multiple outlier detection in multivariate data using projection pursuit techniques

TitleMultiple outlier detection in multivariate data using projection pursuit techniques
Authors
KeywordsDimension reduction
Discordant outliers
Gaussian stochastic process
Multivariate data
Outlier identifier
Primary 62H12
Projection pursuit
Quasi-Monte Carlo methods
Secondary 62A10
Statistical diagnostics
Issue Date2000
PublisherElsevier BV. The Journal's web site is located at http://www.elsevier.com/locate/jspi
Citation
Journal Of Statistical Planning And Inference, 2000, v. 83 n. 1, p. 153-167 How to Cite?
AbstractUsing projection pursuit techniques, in this paper we propose a procedure to detect multiple outliers in multivariate data. The basic idea behind this procedure is to project the multivariate data to univariate observations and then to apply an appropriate univariate outlier identifier to the projected data. The projected outlier identifier forms a centered Gaussian process on the high-dimensional unit sphere. When a set of directions is generated on the unit sphere, the projected outlier identifier on these directions then follows a multivariate normal distribution. In this way, an outlier identifier in the multivariate data with χ2-distribution is constructed. In order to have the outlier identifier revealing much information on multivariate outliers, the directions should be scattered uniformly on the unit sphere as much as possible, which can be implemented in terms of the quasi-Monte Carlo methods. For illustration, three practical data sets are analyzed and compared with existing methods. Also, a simulation is conducted to study the null properties of the multivariate outlier identifier. © 2000 Elsevier Science B.V.
Persistent Identifierhttp://hdl.handle.net/10722/82720
ISSN
2023 Impact Factor: 0.8
2023 SCImago Journal Rankings: 0.736
ISI Accession Number ID
References

 

DC FieldValueLanguage
dc.contributor.authorPan, JXen_HK
dc.contributor.authorFung, WKen_HK
dc.contributor.authorFang, KTen_HK
dc.date.accessioned2010-09-06T08:32:38Z-
dc.date.available2010-09-06T08:32:38Z-
dc.date.issued2000en_HK
dc.identifier.citationJournal Of Statistical Planning And Inference, 2000, v. 83 n. 1, p. 153-167en_HK
dc.identifier.issn0378-3758en_HK
dc.identifier.urihttp://hdl.handle.net/10722/82720-
dc.description.abstractUsing projection pursuit techniques, in this paper we propose a procedure to detect multiple outliers in multivariate data. The basic idea behind this procedure is to project the multivariate data to univariate observations and then to apply an appropriate univariate outlier identifier to the projected data. The projected outlier identifier forms a centered Gaussian process on the high-dimensional unit sphere. When a set of directions is generated on the unit sphere, the projected outlier identifier on these directions then follows a multivariate normal distribution. In this way, an outlier identifier in the multivariate data with χ2-distribution is constructed. In order to have the outlier identifier revealing much information on multivariate outliers, the directions should be scattered uniformly on the unit sphere as much as possible, which can be implemented in terms of the quasi-Monte Carlo methods. For illustration, three practical data sets are analyzed and compared with existing methods. Also, a simulation is conducted to study the null properties of the multivariate outlier identifier. © 2000 Elsevier Science B.V.en_HK
dc.languageengen_HK
dc.publisherElsevier BV. The Journal's web site is located at http://www.elsevier.com/locate/jspien_HK
dc.relation.ispartofJournal of Statistical Planning and Inferenceen_HK
dc.rightsJournal of Statistical Planning and Inference. Copyright © Elsevier BV.en_HK
dc.subjectDimension reductionen_HK
dc.subjectDiscordant outliersen_HK
dc.subjectGaussian stochastic processen_HK
dc.subjectMultivariate dataen_HK
dc.subjectOutlier identifieren_HK
dc.subjectPrimary 62H12en_HK
dc.subjectProjection pursuiten_HK
dc.subjectQuasi-Monte Carlo methodsen_HK
dc.subjectSecondary 62A10en_HK
dc.subjectStatistical diagnosticsen_HK
dc.titleMultiple outlier detection in multivariate data using projection pursuit techniquesen_HK
dc.typeArticleen_HK
dc.identifier.openurlhttp://library.hku.hk:4550/resserv?sid=HKU:IR&issn=0378-3758&volume=83&spage=153&epage=167&date=2000&atitle=Multiple+outlier+detection+in+multivariate+data+using+projection+pursuit+techniquesen_HK
dc.identifier.emailFung, WK: wingfung@hku.hken_HK
dc.identifier.authorityFung, WK=rp00696en_HK
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.scopuseid_2-s2.0-0009872147en_HK
dc.identifier.hkuros55961en_HK
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-0009872147&selection=ref&src=s&origin=recordpageen_HK
dc.identifier.volume83en_HK
dc.identifier.issue1en_HK
dc.identifier.spage153en_HK
dc.identifier.epage167en_HK
dc.identifier.isiWOS:000084148800011-
dc.publisher.placeNetherlandsen_HK
dc.identifier.scopusauthoridPan, JX=7404098188en_HK
dc.identifier.scopusauthoridFung, WK=13310399400en_HK
dc.identifier.scopusauthoridFang, KT=7102880697en_HK
dc.identifier.issnl0378-3758-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats