Adaptive simultaneous confidence intervals in non-parametric estimation

We present non-linear wavelet methods to compute simultaneous confidence intervals for f(x) when f is a functional parameter issued from a non-parametric model. The levels of the intervals are at least [gamma], and we prove that they achieve the minimum diameter up to a logarithmic term. The procedure is data-driven and the adaptation is made via the Lepskii's algorithm.

MoreLess

Year of publication:	2004
Authors:	Tribouley, Karine
Published in:	Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 69.2004, 1, p. 37-51
Publisher:	Elsevier
Subject:	Adaptation Non-parametric estimation-wavelet methods

More details

Type of publication:	Article
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005223742