Full metadata record
DC FieldValueLanguage
dc.contributor.authorHallin, Marc-
dc.contributor.authorPaindaveine, Davy-
dc.contributor.authorVerdebout, Thomas-
dc.date.accessioned2013-04-08T13:35:39Z-
dc.date.available2013-04-08T13:35:39Z-
dc.date.issued2013-04-08-
dc.identifier.urihttp://hdl.handle.net/2003/30135-
dc.identifier.urihttp://dx.doi.org/10.17877/DE290R-5362-
dc.description.abstractWe propose rank-based estimators of principal components, both in the one- sample and, under the assumption of common principal components, in the m- sample cases. Those estimators are obtained via a rank-based version of Le Cam's one-step method, combined with an estimation of cross-information quantities. Under arbitrary elliptical distributions with, in the m-sample case, possibly heterogeneous radial densities, those R-estimators remain root-n consistent and asymptotically normal, while achieving asymptotic e ciency under correctly speci ed densities. Contrary to their traditional counterparts computed from empirical covariances, they do not require any moment conditions. When based on Gaussian score functions, in the one-sample case, they moreover uniformly dominate their classical competitors in the Pitman sense. Their finite-sample performances are investigated via a Monte-Carlo study.en
dc.language.isoende
dc.relation.ispartofseriesDiscussion Paper / SFB 823;12/2013-
dc.subjectcommon principal componentsen
dc.subjectelliptical densitiesen
dc.subjectprincipal componentsen
dc.subjectranksen
dc.subjectR-estimationen
dc.subjectrobustnessen
dc.subjectuniform local asymptotic normalityen
dc.subject.ddc310-
dc.subject.ddc330-
dc.subject.ddc620-
dc.titleEfficient R-estimation of principal and common principal componentsen
dc.typeTextde
dc.type.publicationtypeworkingPaperde
dcterms.accessRightsopen access-
Appears in Collections:Sonderforschungsbereich (SFB) 823

Files in This Item:
File Description SizeFormat 
DP_1213_SFB823_Hallin_Paindaveine_Verdebout.pdfDNB1.31 MBAdobe PDFView/Open


This item is protected by original copyright



This item is protected by original copyright rightsstatements.org