Skip to content

Tous les séminaires

Jeudi 1er Octobre 2009

"Inference on a Generalized Roy Model with Exclusion Restrictions"

de 14 h à 15 h 30 en salle S016
à l'INSEE-CREST, 15 Boulevard Gabriel Péri, 92245 MALAKOFF (Métro : Malakoff/Plateau de Vanves (Immeuble "Malakoff 2)).

Résumé : This paper considers nonparametric identification and estimation of an extension of Roy's model (1951) of occupational choice, which includes a non-pecuniary component in the decision equation and allows for uncertainty on the potential outcomes. This framework is well suited to various economic contexts, including educational and sectoral choices, or migration decisions. We obtain identification through exclusion restrictions, without assuming that the effects of covariates on earnings have large supports. We focus on the identification of the non-pecuniary component when at least one variable affects the selection probability only through potential earnings, that is under the opposite of the usual exclusion restrictions used to identify switching regressions models. Point identification is obtained if such variables are continuous, while bounds are obtained otherwise. We propose a three-stages semiparametric estimation procedure for this model, which yield root-n consistent and asymptotically normal estimators. We also provide finite sample properties relying on Monte Carlo simulations. We finally apply our results to the educational context, by providing new evidence on the influence of non-pecuniary factors on the decision to attend higher education.