English

An Affine Invariant $k$-Nearest Neighbor Regression Estimate

Statistics Theory 2012-05-23 v2 Statistics Theory

Abstract

We design a data-dependent metric in Rd\mathbb R^d and use it to define the kk-nearest neighbors of a given point. Our metric is invariant under all affine transformations. We show that, with this metric, the standard kk-nearest neighbor regression estimate is asymptotically consistent under the usual conditions on kk, and minimal requirements on the input data.

Keywords

Cite

@article{arxiv.1201.0586,
  title  = {An Affine Invariant $k$-Nearest Neighbor Regression Estimate},
  author = {Gérard Biau and Luc Devroye and Vida Dujmovic and Adam Krzyzak},
  journal= {arXiv preprint arXiv:1201.0586},
  year   = {2012}
}
R2 v1 2026-06-21T19:59:28.246Z