Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/22248
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSimar, Léopolden_US
dc.date.accessioned2009-01-29T14:55:06Z-
dc.date.available2009-01-29T14:55:06Z-
dc.date.issued2003en_US
dc.identifier.piurn:nbn:de:kobv:11-10050382-
dc.identifier.urihttp://hdl.handle.net/10419/22248-
dc.description.abstractIn frontier analysis, most of the nonparametric approaches (DEA, FDH) arebased on envelopment ideas which suppose that with probability one, all theobserved units belong to the attainable set. In these "deterministic" frontiermodels, statistical theory is now mostly available. In the presence of noise,this is no more true and envelopment estimators could behave dramaticallysince they are very sensitive to extreme observations that could result onlyfrom noise. DEA/FDH techniques would provide estimators with an error ofthe order of the standard deviation of the noise. In this paper we proposeto adapt some recent results on detecting change points, to improve theperformances of the classical DEA/FDH estimators in the presence of noise.We show by simulated examples that the procedure works well when thenoise is of moderate size, in term of noise to signal ratio. It turns out thatthe procedure is also robust to outliers.en_US
dc.language.isoengen_US
dc.publisheren_US
dc.relation.ispartofseries|aDiscussion papers of interdisciplinary research project 373 |x2003,33en_US
dc.subject.ddc330en_US
dc.subject.keywordNonparametric frontieren_US
dc.subject.keywordStochastic DEA/FDHen_US
dc.subject.keywordRobustness to outliersen_US
dc.subject.stwNichtparametrisches Verfahrenen_US
dc.subject.stwData-Envelopment-Analyseen_US
dc.subject.stwSchätztheorieen_US
dc.subject.stwTheorieen_US
dc.titleHow to Improve the Performances of DEA/FDH Estimators in the Presence of Noise?en_US
dc.typeWorking Paperen_US
dc.identifier.ppn379165767en_US
dc.rightshttp://www.econstor.eu/dspace/Nutzungsbedingungen-
dc.identifier.repecRePEc:zbw:sfb373:200333-

Files in This Item:
File
Size
301.96 kB





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