Lavergne, PascalThomas, A.Universidad Carlos III de Madrid. Departamento de Estadística2009-12-232009-12-231997-09http://hdl.handle.net/10016/6221We propose a flexible framework for estimating and testing structural models with adverse selection. This framework uses semiparametric methods for estimating consistently structural parameters of interest and for assesssing the results by testing procedures. We consider a problem of environmental regulation where firms are regulated through contracts. We show how to check parametric assumptions for the abatement cost function and test for neglected adverse selection. We then apply a semiparametric procedure for estimating models with adverse selection, that does not require to specify the distribution of the private information and avoids costly numerical procedures. The proposed framework can prove useful in a wide variety of problems where adverse selection can be present.application/pdfengAtribución-NoComercial-SinDerivadas 3.0 EspañaSemiparametric estimationspecification testingmodels of adverse selectionenvironmental regulationworking paperEstadísticaopen access