On the efficiency of Gini's mean difference

Carina Gerstenberger (Corresponding Author), Daniel Vogel

Research output: Contribution to journalArticle

12 Citations (Scopus)
1 Downloads (Pure)

Abstract

We examine the efficiency of the mean deviation and Gini's mean difference (the mean of all pairwise distances). Our findings support the viewpoint that Gini's mean difference combines the advantages of the mean deviation and the standard deviation.
Original languageEnglish
Pages (from-to)569-596
Number of pages28
JournalStatistical Methods & Applications
Volume24
Issue number4
Early online date7 May 2015
DOIs
Publication statusPublished - Nov 2015

Fingerprint

Mean deviation
Standard deviation
Pairwise
Deviation
Gini

Keywords

  • math.ST
  • stat.TH
  • 62G35, 62G05, 62G20
  • influence function
  • mean deviation
  • median absolute deviation
  • normal mixture distribution
  • residue theorem
  • robustness

Cite this

On the efficiency of Gini's mean difference. / Gerstenberger, Carina (Corresponding Author); Vogel, Daniel.

In: Statistical Methods & Applications, Vol. 24, No. 4, 11.2015, p. 569-596.

Research output: Contribution to journalArticle

Gerstenberger, Carina ; Vogel, Daniel. / On the efficiency of Gini's mean difference. In: Statistical Methods & Applications. 2015 ; Vol. 24, No. 4. pp. 569-596.
@article{955ac8ba30e4462196f7f57d1d654a4e,
title = "On the efficiency of Gini's mean difference",
abstract = "We examine the efficiency of the mean deviation and Gini's mean difference (the mean of all pairwise distances). Our findings support the viewpoint that Gini's mean difference combines the advantages of the mean deviation and the standard deviation.",
keywords = "math.ST, stat.TH, 62G35, 62G05, 62G20, influence function, mean deviation, median absolute deviation, normal mixture distribution, residue theorem, robustness",
author = "Carina Gerstenberger and Daniel Vogel",
note = "18 pages, 3 figures, 8 tables Acknowledgments We are indebted to Herold Dehling for introducing us to the theory of U-statistics, to Roland Fried for introducing us to robust statistics, and to Alexander D{\"u}rre, who has demonstrated the benefit of complex analysis for solving statistical problems. Both authors were supported in part by the Collaborative Research Centre 823 Statistical modelling of nonlinear dynamic processes.",
year = "2015",
month = "11",
doi = "10.1007/s10260-015-0315-x",
language = "English",
volume = "24",
pages = "569--596",
journal = "Statistical Methods & Applications",
issn = "1618-2510",
publisher = "Physica-Verlag",
number = "4",

}

TY - JOUR

T1 - On the efficiency of Gini's mean difference

AU - Gerstenberger, Carina

AU - Vogel, Daniel

N1 - 18 pages, 3 figures, 8 tables Acknowledgments We are indebted to Herold Dehling for introducing us to the theory of U-statistics, to Roland Fried for introducing us to robust statistics, and to Alexander Dürre, who has demonstrated the benefit of complex analysis for solving statistical problems. Both authors were supported in part by the Collaborative Research Centre 823 Statistical modelling of nonlinear dynamic processes.

PY - 2015/11

Y1 - 2015/11

N2 - We examine the efficiency of the mean deviation and Gini's mean difference (the mean of all pairwise distances). Our findings support the viewpoint that Gini's mean difference combines the advantages of the mean deviation and the standard deviation.

AB - We examine the efficiency of the mean deviation and Gini's mean difference (the mean of all pairwise distances). Our findings support the viewpoint that Gini's mean difference combines the advantages of the mean deviation and the standard deviation.

KW - math.ST

KW - stat.TH

KW - 62G35, 62G05, 62G20

KW - influence function

KW - mean deviation

KW - median absolute deviation

KW - normal mixture distribution

KW - residue theorem

KW - robustness

U2 - 10.1007/s10260-015-0315-x

DO - 10.1007/s10260-015-0315-x

M3 - Article

VL - 24

SP - 569

EP - 596

JO - Statistical Methods & Applications

JF - Statistical Methods & Applications

SN - 1618-2510

IS - 4

ER -