File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Supervised singular value decomposition and its asymptotic properties

TitleSupervised singular value decomposition and its asymptotic properties
Authors
KeywordsPrincipal component analysis
Low rank approximation
62H12
SupSVD
Supervised dimension reduction
Reduced rank regression
Issue Date2016
Citation
Journal of Multivariate Analysis, 2016, v. 146, p. 7-17 How to Cite?
Abstract© 2015 Elsevier Inc. A supervised singular value decomposition (SupSVD) model has been developed for supervised dimension reduction where the low rank structure of the data of interest is potentially driven by additional variables measured on the same set of samples. The SupSVD model can make use of the information in the additional variables to accurately extract underlying structures that are more interpretable. The model is general and includes the principal component analysis model and the reduced rank regression model as two extreme cases. The model is formulated in a hierarchical fashion using latent variables, and a modified expectation-maximization algorithm for parameter estimation is developed, which is computationally efficient. The asymptotic properties for the estimated parameters are derived. We use comprehensive simulations and a real data example to illustrate the advantages of the SupSVD model.
Persistent Identifierhttp://hdl.handle.net/10722/219833
ISSN
2015 Impact Factor: 0.857
2015 SCImago Journal Rankings: 1.458

 

DC FieldValueLanguage
dc.contributor.authorLi, Gen-
dc.contributor.authorYang, Dan-
dc.contributor.authorNobel, Andrew B.-
dc.contributor.authorShen, Haipeng-
dc.date.accessioned2015-09-23T02:58:03Z-
dc.date.available2015-09-23T02:58:03Z-
dc.date.issued2016-
dc.identifier.citationJournal of Multivariate Analysis, 2016, v. 146, p. 7-17-
dc.identifier.issn0047-259X-
dc.identifier.urihttp://hdl.handle.net/10722/219833-
dc.description.abstract© 2015 Elsevier Inc. A supervised singular value decomposition (SupSVD) model has been developed for supervised dimension reduction where the low rank structure of the data of interest is potentially driven by additional variables measured on the same set of samples. The SupSVD model can make use of the information in the additional variables to accurately extract underlying structures that are more interpretable. The model is general and includes the principal component analysis model and the reduced rank regression model as two extreme cases. The model is formulated in a hierarchical fashion using latent variables, and a modified expectation-maximization algorithm for parameter estimation is developed, which is computationally efficient. The asymptotic properties for the estimated parameters are derived. We use comprehensive simulations and a real data example to illustrate the advantages of the SupSVD model.-
dc.languageeng-
dc.relation.ispartofJournal of Multivariate Analysis-
dc.subjectPrincipal component analysis-
dc.subjectLow rank approximation-
dc.subject62H12-
dc.subjectSupSVD-
dc.subjectSupervised dimension reduction-
dc.subjectReduced rank regression-
dc.titleSupervised singular value decomposition and its asymptotic properties-
dc.typeArticle-
dc.description.natureLink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.jmva.2015.02.016-
dc.identifier.scopuseid_2-s2.0-84925707082-
dc.identifier.hkuros263859-
dc.identifier.volume146-
dc.identifier.spage7-
dc.identifier.epage17-
dc.identifier.eissn1095-7243-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats