103
33

Adaptive estimation of a distribution function and its density in sup-norm loss by wavelet and spline projections

Abstract

Given an i.i.d. sample from a distribution FF on R\mathbb{R} with uniformly continuous density p0p_0, purely data-driven estimators are constructed that efficiently estimate FF in sup-norm loss and simultaneously estimate p0p_0 at the best possible rate of convergence over H\"older balls, also in sup-norm loss. The estimators are obtained by applying a model selection procedure close to Lepski's method with random thresholds to projections of the empirical measure onto spaces spanned by wavelets or BB-splines. The random thresholds are based on suprema of Rademacher processes indexed by wavelet or spline projection kernels. This requires Bernstein-type analogs of the inequalities in Koltchinskii [Ann. Statist. 34 (2006) 2593-2656] for the deviation of suprema of empirical processes from their Rademacher symmetrizations.

View on arXiv
Comments on this paper