EconStor >
Humboldt-Universität Berlin >
Sonderforschungsbereich 649: Ökonomisches Risiko, Humboldt-Universität Berlin >
SFB 649 Discussion Papers, HU Berlin >

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10419/25034
  

Full metadata record

DC FieldValueLanguage
dc.contributor.authorČίžek, Pavelen_US
dc.contributor.authorHärdle, Wolfgang Karlen_US
dc.date.accessioned2005-08-15en_US
dc.date.accessioned2009-07-23T14:42:49Z-
dc.date.available2009-07-23T14:42:49Z-
dc.date.issued2005en_US
dc.identifier.urihttp://hdl.handle.net/10419/25034-
dc.description.abstractMost dimension reduction methods based on nonparametric smoothing are highly sensitive to outliers and to data coming from heavy-tailed distributions. We show that the recently proposed methods by Xia et al. (2002) can be made robust in such a way that preserves all advantages of the original approach. Their extension based on the local one-step M-estimators is sufficiently robust to outliers and data from heavy tailed distributions, it is relatively easy to implement, and surprisingly, it performs as well as the original methods when applied to normally distributed data.en_US
dc.language.isoengen_US
dc.publisherSFB 649, Economic Risk Berlinen_US
dc.relation.ispartofseriesSFB 649 discussion paper 2005,015en_US
dc.subject.ddc330en_US
dc.subject.keywordDimension reductionen_US
dc.subject.keywordNonparametric regressionen_US
dc.subject.keywordM-estimationen_US
dc.subject.stwNichtparametrisches Verfahrenen_US
dc.subject.stwRobustes Verfahrenen_US
dc.subject.stwSchätztheorieen_US
dc.subject.stwTheorieen_US
dc.titleRobust estimation of dimension reduction spaceen_US
dc.typeWorking Paperen_US
dc.identifier.ppn496003585en_US
dc.rightshttp://www.econstor.eu/dspace/Nutzungsbedingungen-
Appears in Collections:SFB 649 Discussion Papers, HU Berlin

Files in This Item:
File Description SizeFormat
496003585.PDF329.58 kBAdobe PDF
No. of Downloads: Counter Stats
Show simple item record
Download bibliographical data as: BibTeX

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.