November 15th 2011
Faculty of Economic and Business Sciences | Seminar 273
Estimation of the error distribution in a semiparametric transformation model
2011/11/15 – 12:30 h | Cédric Heuchenne, University of Liége and Université Catholique de Louvain.
Abstract
In this paper, we consider the semiparametric transformation model \Lambda_{\theta_0}(Y) = m(X) + u, where \theta_0 is an unknown finite dimensional parameter, the function m(.) = E[ \Lambda_{\theta_0}(Y ) / X =. ] is smooth but otherwise unknown, and the covariate X is independent of the error u. Estimators for the density and the cumulative distribution functions of u are investigated and their convergence properties are proved.
The proposed estimators depend on a profile likelihood estimator of \theta_0 and a nonparametric kernel estimator of m. We also evaluate the practical performance of our estimators in a simulation study for several models and sample sizes.
Joint work with Rawane Samb and Ingrid van Keilegom.