On the mutual nearest neighbors estimate in regression - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles Journal of Machine Learning Research Year : 2013

On the mutual nearest neighbors estimate in regression

Abstract

Motivated by promising experimental results, this paper investigates the theoretical properties of a recently proposed nonparametric estimator, called the Mutual Nearest Neighbors rule, which estimates the regression function m(x) = E[Y vertical bar X = x] as follows: first identify the k nearest neighbors of x in the sample D-n, then keep only those for which x is itself one of the k nearest neighbors, and finally take the average over the corresponding response variables. We prove that this estimator is consistent and that its rate of convergence is optimal. Since the estimate with the optimal rate of convergence depends on the unknown distribution of the observations, we also present adaptation results by data-splitting.
No file

Dates and versions

hal-00874050 , version 1 (17-10-2013)

Identifiers

  • HAL Id : hal-00874050 , version 1

Cite

Arnaud Guyader, Nick Hengartner. On the mutual nearest neighbors estimate in regression. Journal of Machine Learning Research, 2013, 14, pp.2361-2376. ⟨hal-00874050⟩
116 View
0 Download

Share

Gmail Facebook X LinkedIn More