Multiplicative Bias Corrected Nonparametric Smoothers with Application to Nuclear Energy Spectrum Estimation

The paper presents a multiplicative bias reduction estimator for non-parametric regression. The approach consists to apply a multiplicative bias correction to an oversmooth pilot estimator. We study the asymp-totic properties of the resulting estimate and prove that this estimate has zero asymptotic bias and the same asymptotic variance as the lo-cal linear estimate. Simulations show that our asymptotic results are available for small sample sizes. We also illustrate the benefit of this new method on nuclear energy spectrum estimation. Index terms: Nonparametric regression, bias reduction, local linear estimate.