Residual variance estimation using a nearest neighbor statistic

Elia Liitiäinen*, Francesco Corona, Amaury Lendasse

*Corresponding author for this work

    Research output: Contribution to journalArticleScientificpeer-review

    20 Citations (Scopus)


    In this paper we consider the problem of estimating E [(Y - E [Y {divides} X])2] based on a finite sample of independent, but not necessarily identically distributed, random variables (Xi, Yi)i = 1M. We analyze the theoretical properties of a recently developed estimator. It is shown that the estimator has many theoretically interesting properties, while the practical implementation is simple.

    Original languageEnglish
    Pages (from-to)811-823
    Number of pages13
    JournalJournal of Multivariate Analysis
    Issue number4
    Publication statusPublished - 1 Apr 2010
    MoE publication typeA1 Journal article-refereed


    • Nearest neighbors
    • Noise variance estimation
    • Nonparametric
    • Residual variance estimation


    Dive into the research topics of 'Residual variance estimation using a nearest neighbor statistic'. Together they form a unique fingerprint.

    Cite this