Model selection for density estimation with L2-loss

Abstract

We consider here estimation of an unknown probability density s belonging to L2(mu) where mu is a probability measure. We have at hand n i.i.d. observations with density s and use the squared L2-norm as our loss function. The purpose of this paper is to provide an abstract but completely general method for estimating s by model selection, allowing to handle arbitrary families of finite-dimensional (possibly non-linear) models and any density s belonging to L2(mu). We shall, in particular, consider the cases of unbounded densities and bounded densities with unknown bound and investigate how the L-infinity-norm of s may influence the risk. We shall also provide applications to adaptive estimation and aggregation of preliminary estimators. Although of a purely theoretical nature, our method leads to results that cannot presently be reached by more concrete methods.

37 pages. Minor changes

References in corpus (2)

• Adaptive estimation of and oracle inequalities for probability densities and characteristic functions
• Model selection for Poisson processes

Cited by in corpus (11)

• Multivariate convex regression: global risk bounds and adaptation
• Adaptation to lowest density regions with application to support recovery
• Optimal model selection for density estimation of stationary data under various mixing conditions
• Optimal model selection in density estimation
• Dependence structure estimation using Copula Recursive Trees
• Robust estimation on a parametric model via testing
• An efficient algorithm for T-estimation
• On adaptive minimax density estimation on $R^d$
• Adaptive directional estimator of the density in R^d for independent and mixing sequences
• Estimator selection with respect to Hellinger-type risks
• Estimation in the convolution structure density model. Part II: adaptation over the scale of anisotropic classes