Tous les séminaires
Jeudi 24 Septembre 2009
Laurent DAVEZIES (INSEE)
"Effet redistributif du système d'assurance maladie : Une approche par micro simulation sur cycle de vie"
avec Valérie ALBOUY et Thierry DEBRAND
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 : In this article, we focus on the estimation of outpatient expenditures with panel data. Facing skewed distributions, we model the logarithm of expenditures and consider five different models. The first two are cross section two part and sample selection models. We establish that panel data two part models are unrealistic in our setting. Thus, we estimate three sample selection models in panel data: a static one and two dynamic ones. These two latter differ in their assumptions on the variance of the residual. Modeling heteroskedasticity may indeed be important to avoid the bias due to the retransformation problem. We show that lagged endogenous are important factors of heteroskedasticity. For the models with state dependence we provide a new solution to the initial conditions problem by controlling for generalized residuals. We establish that panel data models highly improve the correlation explained by the model in the time-series dimension without damaging the fit in the cross-section dimension. For all indicators of fit, the model with state dependence and heteroskedasticity seems to dominate the others.