Universal Adaptive Estimations and Confidence Intervals in the Nonparametric Statistics

Abstract

The paper considers so-called adaptive estimations of regression, distribution density and spectral density of a Gaussian stationary sequence, asymptotically optimal in order at a growing number of observation on any regular subspace compactly embedded in space $L_2$, and confidence intervals, also adaptive, are constructed on their basis for the estimated functions in an integral norm.