A clustering law for some discrete order statistics

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Let X1, X2, ..., Xn be a sequence of independent, identically distributed positive integer random variables with distribution function F. Anderson (1970) proved a variant of the law of large numbers by showing that the sample maximum moves asymptotically on two values if and only if F satisfies a 'clustering' condition, limn → ∞[1 -F(n + 1)]/[1 - F(n)] = 0. In this article, we generalize Anderson's result and show that it is robust by proving that, for any r ≥ 0, the sample maximum and other extremes asymptotically cluster on r + 2 values if and only if limn → ∞ [1 - F(n + r + 1)]/[1 - F(n)] = 0. Together with previous work which considered other asymptotic properties of these sample extremes, a more detailed asymptotic clustering structure for discrete order statistics is presented.

Original languageEnglish (US)
Pages (from-to)226-241
Number of pages16
JournalJournal of Applied Probability
Volume40
Issue number1
DOIs
StatePublished - Mar 2003
Externally publishedYes

Fingerprint

Order Statistics
Clustering
Extremes
If and only if
Law of large numbers
Identically distributed
Asymptotic Properties
Distribution Function
Random variable
Generalise
Integer
Order statistics

Keywords

  • Asymptotics
  • Clustering
  • Extremes
  • Law of large numbers
  • Order statistics

ASJC Scopus subject areas

  • Mathematics(all)
  • Statistics and Probability

Cite this

A clustering law for some discrete order statistics. / Sethuraman, Sunder.

In: Journal of Applied Probability, Vol. 40, No. 1, 03.2003, p. 226-241.

Research output: Contribution to journalArticle

@article{8f1313d48dff4c11894e2a991af6ba7d,
title = "A clustering law for some discrete order statistics",
abstract = "Let X1, X2, ..., Xn be a sequence of independent, identically distributed positive integer random variables with distribution function F. Anderson (1970) proved a variant of the law of large numbers by showing that the sample maximum moves asymptotically on two values if and only if F satisfies a 'clustering' condition, limn → ∞[1 -F(n + 1)]/[1 - F(n)] = 0. In this article, we generalize Anderson's result and show that it is robust by proving that, for any r ≥ 0, the sample maximum and other extremes asymptotically cluster on r + 2 values if and only if limn → ∞ [1 - F(n + r + 1)]/[1 - F(n)] = 0. Together with previous work which considered other asymptotic properties of these sample extremes, a more detailed asymptotic clustering structure for discrete order statistics is presented.",
keywords = "Asymptotics, Clustering, Extremes, Law of large numbers, Order statistics",
author = "Sunder Sethuraman",
year = "2003",
month = "3",
doi = "10.1239/jap/1044476836",
language = "English (US)",
volume = "40",
pages = "226--241",
journal = "Journal of Applied Probability",
issn = "0021-9002",
publisher = "University of Sheffield",
number = "1",

}

TY - JOUR

T1 - A clustering law for some discrete order statistics

AU - Sethuraman, Sunder

PY - 2003/3

Y1 - 2003/3

N2 - Let X1, X2, ..., Xn be a sequence of independent, identically distributed positive integer random variables with distribution function F. Anderson (1970) proved a variant of the law of large numbers by showing that the sample maximum moves asymptotically on two values if and only if F satisfies a 'clustering' condition, limn → ∞[1 -F(n + 1)]/[1 - F(n)] = 0. In this article, we generalize Anderson's result and show that it is robust by proving that, for any r ≥ 0, the sample maximum and other extremes asymptotically cluster on r + 2 values if and only if limn → ∞ [1 - F(n + r + 1)]/[1 - F(n)] = 0. Together with previous work which considered other asymptotic properties of these sample extremes, a more detailed asymptotic clustering structure for discrete order statistics is presented.

AB - Let X1, X2, ..., Xn be a sequence of independent, identically distributed positive integer random variables with distribution function F. Anderson (1970) proved a variant of the law of large numbers by showing that the sample maximum moves asymptotically on two values if and only if F satisfies a 'clustering' condition, limn → ∞[1 -F(n + 1)]/[1 - F(n)] = 0. In this article, we generalize Anderson's result and show that it is robust by proving that, for any r ≥ 0, the sample maximum and other extremes asymptotically cluster on r + 2 values if and only if limn → ∞ [1 - F(n + r + 1)]/[1 - F(n)] = 0. Together with previous work which considered other asymptotic properties of these sample extremes, a more detailed asymptotic clustering structure for discrete order statistics is presented.

KW - Asymptotics

KW - Clustering

KW - Extremes

KW - Law of large numbers

KW - Order statistics

UR - http://www.scopus.com/inward/record.url?scp=0037573234&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037573234&partnerID=8YFLogxK

U2 - 10.1239/jap/1044476836

DO - 10.1239/jap/1044476836

M3 - Article

AN - SCOPUS:0037573234

VL - 40

SP - 226

EP - 241

JO - Journal of Applied Probability

JF - Journal of Applied Probability

SN - 0021-9002

IS - 1

ER -