64
80

Uniform limit theorems for wavelet density estimators

Abstract

Let pn(y)=kα^kϕ(yk)+l=0jn1kβ^lk2l/2ψ(2lyk)p_n(y)=\sum_k\hat{\alpha}_k\phi(y-k)+\sum_{l=0}^{j_n-1}\sum_k\hat {\beta}_{lk}2^{l/2}\psi(2^ly-k) be the linear wavelet density estimator, where ϕ\phi, ψ\psi are a father and a mother wavelet (with compact support), α^k\hat{\alpha}_k, β^lk\hat{\beta}_{lk} are the empirical wavelet coefficients based on an i.i.d. sample of random variables distributed according to a density p0p_0 on R\mathbb{R}, and jnZj_n\in\mathbb{Z}, jnj_n\nearrow\infty. Several uniform limit theorems are proved: First, the almost sure rate of convergence of supyRpn(y)Epn(y)\sup_{y\in\mathbb{R}}|p_n(y)-Ep_n(y)| is obtained, and a law of the logarithm for a suitably scaled version of this quantity is established. This implies that supyRpn(y)p0(y)\sup_{y\in\mathbb{R}}|p_n(y)-p_0(y)| attains the optimal almost sure rate of convergence for estimating p0p_0, if jnj_n is suitably chosen. Second, a uniform central limit theorem as well as strong invariance principles for the distribution function of pnp_n, that is, for the stochastic processes n(FnW(s)F(s))=ns(pnp0),sR\sqrt{n}(F_n ^W(s)-F(s))=\sqrt{n}\int_{-\infty}^s(p_n-p_0),s\in\mathbb{R}, are proved; and more generally, uniform central limit theorems for the processes n(pnp0)f\sqrt{n}\int(p_n-p_0)f, fFf\in\mathcal{F}, for other Donsker classes F\mathcal{F} of interest are considered. As a statistical application, it is shown that essentially the same limit theorems can be obtained for the hard thresholding wavelet estimator introduced by Donoho et al. [Ann. Statist. 24 (1996) 508--539].

View on arXiv
Comments on this paper