Errors-in-variables models: a generalized functions approach VERSION UNDER REVISION: do not quote

Identi cation in errors-in-variables regression models was extended to wide models classes by S. Schennach (Econometrica, 2007) (S) via use of generalized functions. The problems addressed in (S) are revis-ited to extend and correct some of the results. Nonparametric identi-cation here avoids decomposition of generalized functions into ordinary and singular parts, which may not hold, and gives a corrected proof. Continuity of the identi\u85cation mapping and possible nonparametric estimation is discussed. Semiparametric identi\u85cation via a \u85nite set of moment conditions holds for classes of parametric functions; here such functions are explicitely characterized, the restriction in (S) that a moment generating function exist for the measurement error distri-bution is avoided; corrected weighting functions are provided