Humboldt University of Berlin, Interdisciplinary Research Project 373: Quantification and Simulation of Economic Processes, Berlin
We consider a problem of estimation of parametric component in a partial linear model. Suppose that a finite set E of linear estimators is given. Our goal is to mimic the estimator in E that has the smallest risk. Using a second order expansion of the risk of linear estimators we propose a practically feasible adaptive procedure for choke of smoothing parameters based on the principle of unbiased risk estimation.