Optimal model selection in density estimation

Abstract

We build penalized least-squares estimators using the slope heuristic and resampling penalties. We prove oracle inequalities for the selected estimator with leading constant asymptotically equal to 1. We compare the practical performances of these methods in a short simulation study.

(minor revision)

References in corpus (4)

• Concentration around the mean for maxima of empirical processes
• Data-driven calibration of penalties for least-squares regression
• Model selection by resampling penalization
• Model selection for density estimation with L2-loss

Cited by in corpus (2)

• Optimal model selection for density estimation of stationary data under various mixing conditions
• Adaptive non-asymptotic confidence balls in density estimation