Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-030-73249-3_4
info:eu-repo/semantics/OpenAccess
Éric Gautier, « Relaxing Monotonicity in Endogenous Selection Models and Application to Surveys », HAL SHS (Sciences de l’Homme et de la Société), ID : 10.1007/978-3-030-73249-3_4
This paper considers endogenous selection models, in particular nonparametric ones. Estimating the unconditional law of the outcomes is possible when one uses instrumental variables. Using a selection equation which is additively separable in a one dimensionalunobservable has the sometimes undesirable property of instrument monotonicity. We present models which allow for nonmonotonicity and are based on nonparametric random coefficients indices. We discuss their non parametric identification and apply these results to inference on nonlinear statistics such as the Gini index in surveys when the nonresponse is not missing atrandom.