A new algorithm for estimating the effective dimension-reduction subspace - Archive ouverte HAL Access content directly
Journal Articles Journal of Machine Learning Research Year : 2008

A new algorithm for estimating the effective dimension-reduction subspace

Abstract

The statistical problem of estimating the effective dimension-reduction (EDR) subspace in the multi-index regression model with deterministic design and additive noise is considered. A new procedure for recovering the directions of the EDR subspace is proposed. Under mild assumptions, $\sqrt n$-consistency of the proposed procedure is proved (up to a logarithmic factor) in the case when the structural dimension is not larger than $4$. The empirical behavior of the algorithm is studied through numerical simulations.
Fichier principal
Vignette du fichier
EDR-EST_JMLR.pdf (298.37 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00128129 , version 1 (30-01-2007)
hal-00128129 , version 2 (18-04-2007)

Identifiers

Cite

Arnak S. Dalalyan, Anatoli B. Juditsky, Vladimir Spokoiny. A new algorithm for estimating the effective dimension-reduction subspace. Journal of Machine Learning Research, 2008, 9, pp.1647-1678. ⟨hal-00128129v2⟩
216 View
165 Download

Altmetric

Share

Gmail Facebook X LinkedIn More