Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/64684 
Full metadata record
DC FieldValueLanguage
dc.contributor.authorHahn, Jinyongen
dc.contributor.authorRidder, Geerten
dc.date.accessioned2010-10-07-
dc.date.accessioned2012-10-16T13:12:13Z-
dc.date.available2012-10-16T13:12:13Z-
dc.date.issued2010-
dc.identifier.pidoi:10.1920/wp.cem.2010.2310en
dc.identifier.urihttp://hdl.handle.net/10419/64684-
dc.description.abstractWe study the asymptotic distribution of three-step estimators of a finite dimensional parameter vector where the second step consists of one or more nonparametric regressions on a regressor that is estimated in the first step. The first step estimator is either parametric or non-parametric. Using Newey's (1994) path-derivative method we derive the contribution of the first step estimator to the influence function. In this derivation it is important to account for the dual role that the first step estimator plays in the second step non-parametric regression, i.e., that of conditioning variable and that of argument. We consider three examples in more detail: the partial linear regression model estimator with a generated regressor, the Heckman, Ichimura and Todd (1998) estimator of the Average Treatment Effect and a semi-parametric control variable estimator.en
dc.language.isoengen
dc.publisher|aCentre for Microdata Methods and Practice (cemmap) |cLondonen
dc.relation.ispartofseries|acemmap working paper |xCWP23/10en
dc.subject.jelC01en
dc.subject.jelC14en
dc.subject.ddc330en
dc.subject.keywordsemi-parametric estimationen
dc.subject.keywordgenerated regressorsen
dc.subject.keywordasymptotic varianceen
dc.titleThe asymptotic variance of semi-parametric estimators with generated regressors-
dc.typeWorking Paperen
dc.identifier.ppn636320251en
dc.rightshttp://www.econstor.eu/dspace/Nutzungsbedingungenen
dc.identifier.repecRePEc:ifs:cemmap:23/10en

Files in This Item:
File
Size
314.97 kB





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