Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/20231
Full metadata record
DC FieldValueLanguage
dc.contributor.authorPaserman, Marco Danieleen_US
dc.date.accessioned2009-01-28T16:12:35Z-
dc.date.available2009-01-28T16:12:35Z-
dc.date.issued2004en_US
dc.identifier.urihttp://hdl.handle.net/10419/20231-
dc.description.abstractThis paper describes a semiparametric Bayesian method for analyzing duration data. Theproposed estimator specifies a complete functional form for duration spells, but allowsflexibility by introducing an individual heterogeneity term, which follows a Dirichlet mixturedistribution. I show how to obtain predictive distributions for duration data that correctlyaccount for the uncertainty present in the model. I also directly compare the performance ofthe proposed estimator with Heckman and Singer's (1984) Non Parametric MaximumLikelihood Estimator (NPMLE). The methodology is applied to the analysis of youthunemployment spells. Compared to the NPMLE, the proposed estimator reflects moreaccurately the uncertainty surrounding the heterogeneity distribution.en_US
dc.language.isoengen_US
dc.publisher|aInstitute for the Study of Labor (IZA) |cBonn-
dc.relation.ispartofseries|aIZA Discussion paper series |x996en_US
dc.subject.jelC41en_US
dc.subject.jelC11en_US
dc.subject.ddc330en_US
dc.subject.keywordduration dataen_US
dc.subject.keywordDirichlet processen_US
dc.subject.keywordBayesian inferenceen_US
dc.subject.keywordMarkov chain Monte Carlo simulationen_US
dc.subject.stwStatistische Bestandsanalyseen_US
dc.subject.stwNichtparametrisches Verfahrenen_US
dc.subject.stwBayes-Statistiken_US
dc.subject.stwMaximum-Likelihood-Methodeen_US
dc.subject.stwSchätzungen_US
dc.subject.stwJugendarbeitslosigkeiten_US
dc.subject.stwTheorieen_US
dc.subject.stwVereinigte Staatenen_US
dc.titleBayesian inference for duration data with unobserved and unknown heterogeneity : Monte Carlo evidence and an applicationen_US
dc.typeWorking Paperen_US
dc.identifier.ppn378234463en_US
dc.rightshttp://www.econstor.eu/dspace/Nutzungsbedingungen-

Files in This Item:
File
Size
523.58 kB





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