Asymptotic expansions of the probabilities of misclassification for k-NN discriminant rules

Kamal C. Chanda

Statistics & Probability Letters, 1990, vol. 10, issue 4, 341-349

Abstract: Let X1,..., Xl and Y1,..., Yn be independent random samples from the distributions functions (d.f.) F and G respectively. Assume that F' = f and G' = g. The discriminant rule for classifying an independently sampled observation z to F if (z) > (z), and to G otherwise where and are the nearest neighbor estimates of f and g respectively based on the 'training' X- and Y-samples is considered optimal in some sense. Let PF denote the probability measure under the assumption that Z ~ F and set P0 = PF(f(Z) P0 as l, n --> [infinity], for the situation where l = n, F(x) = M(x - [theta]2) and G(x) = M(x - [theta]1) for some symmetric d.f. M with a compact support and parameters [theta]1, [theta]2. It is also established through an example that this rate of convergence does not necessarily hold if the support of the symmetric d.f. M is infinite.

Keywords: Nearest; neighbor; discriminant; rule; probability; of; misclassification; asymptotic; expansions (search for similar items in EconPapers)
Date: 1990
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/0167-7152(90)90053-A
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:stapro:v:10:y:1990:i:4:p:341-349

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/supportfaq.cws_home/regional
https://shop.elsevie ... _01_ooc_1&version=01

Access Statistics for this article

Statistics & Probability Letters is currently edited by Somnath Datta and Hira L. Koul

More articles in Statistics & Probability Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().