Regression with random design: a minimax study
Abstract
The problem of estimating a regression function based on a regression model with (known) random design is considered. By adopting the framework of wavelet analysis, we establish the asymptotic minimax rate of convergence over Besov balls under the Lp risk . A part of this paper is devoted to the case where the design density is vanishing.
Domains
Statistics [math.ST]
Loading...