Abstract: In this paper we propose a kernel-based method for estimating additive partially linear models. Our method makes use of the partially linear model structure at the initial stage when estimating the individual nonparametric components. Monte Carlo simulations show that our proposed estimator performs quite well for moderate sample sizes. In addition, we provide a consistent estimator for the asymptotic variance of the estimator of the parameter in the linear part of the model, where the linear component variable can be discrete or continuous. This facilitates inferential procedures based on our proposed estimator for the finite dimensional parameter. Our result also leads to a simple identification condition for the finite dimensional parameter.
Key words and phrases: Additive partially linear model, finite sample efficiency, identification, kernel smoother, simulations.