English

A new algorithm for estimating the effective dimension-reduction subspace

Statistics Theory 2007-06-13 v1 Statistics Theory

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, n\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.

Keywords

Cite

@article{arxiv.math/0701887,
  title  = {A new algorithm for estimating the effective dimension-reduction subspace},
  author = {Arnak Dalalyan and Anatoly Juditsky and Vladimir Spokoiny},
  journal= {arXiv preprint arXiv:math/0701887},
  year   = {2007}
}