2014StatisticsRequires access

Maximal dispersion of order statistics in dependent samples

Tomasz Rychlik

Open publisher page 4 citations

Abstract

Consider n arbitrarily dependent identically distributed random variables. Rychlik [Extreme variances of order statistics in dependent samples. Statist Probab Lett. 2008;78:1577–1582] proved that the optimal bound for the ratio of variances of the mth order statistic and a single-parent variable is the maximum of n/m and n/(n+1−m). We show here that the evaluation is optimal for many other measures of dispersion of the order statistics and original variable.

About this research paper

What this paper is about

Consider n arbitrarily dependent identically distributed random variables. Rychlik [Extreme variances of order statistics in dependent samples. Statist Probab Lett. 2008;78:1577–1582] proved that the optimal bound for the ratio of variances of the mth order statistic and a single-parent variable is the maximum of n/m and n/(n+1−m). We show here that the evaluation is optimal for many other measures of dispersion of the order statistics and original variable.

Why it matters

OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Consider n arbitrarily dependent identically distributed random variables. Rychlik [Extreme variances of order statistics in dependent samples. Statist Probab Lett. 2008;78:1577–1582] proved that the optimal bound for the ratio of variances of the mth order statistic and a single-parent variable is the maximum of n/m and n/(n+1−m). We show here that the evaluation is optimal for many other measures of dispersion of the order statistics and original variable.

Key concepts: Mathematics, Independent and identically distributed random variables, Order statistic, Statistics, Statistic, Random variable, Dispersion (optics), Order (exchange)

Related papers

Back to paper searchBrowse research topicsOriginal source
Maximal dispersion of order statistics in dependent samples — Research Paper | ScholarLens