A BANDWIDTH SELECTOR FOR BIVARIATE KERNEL REGRESSION

For two and higher dimensional kernel regression, currently available bandwidth selection procedures are based on cross-validation or relate d penalizing ideas. However, these techniques have been shown to suffe r from high sample variability and, in addition, can sometimes be diff icult to implement when a vector of bandwidths needs to be selected. I n this paper we propose a selector based on an iterative plug-in appro ach for bivariate kernel regression. It is shown to give satisfactory results and can be quickly computed. Our ideas can be extended to high er dimensions.