A numerical study of the power function of a new symmetry test

D. Bagkavos*, P. N. Patil, A. T.A. Wood

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

A new nonparametric test for the null hypothesis of symmetry is proposed. A necessary and sufficient condition for symmetry, which is based on the fact that under symmetry the covariance between the probability density and cumulative distribution functions of the underlying population is zero, is used to define the test statistic. The main emphasis here is on the small sample power properties of the test. Through simulations with samples generated from a wide range of distributions, it is shown that the test has a reasonable power function which compares favorably against many other existing tests of symmetry. It is also shown that the defining feature of this test is “the higher the asymmetry higher is the power”.

Original languageEnglish
Title of host publicationNonparametric Statistics - 2nd ISNPS 2014
EditorsJuan Romo, Ricardo Cao, Wenceslao González Manteiga
PublisherSpringer New York LLC
Pages1-10
Number of pages10
ISBN (Print)9783319415819
DOIs
Publication statusPublished - 2016
Externally publishedYes
Event2nd Conference of the International Society for Nonparametric Statistics, ISNPS 2014 - Cadiz, Spain
Duration: 12 Jun 201416 Jun 2014

Publication series

NameSpringer Proceedings in Mathematics and Statistics
Volume175
ISSN (Print)2194-1009
ISSN (Electronic)2194-1017

Conference

Conference2nd Conference of the International Society for Nonparametric Statistics, ISNPS 2014
Country/TerritorySpain
CityCadiz
Period12/06/1416/06/14

Fingerprint

Dive into the research topics of 'A numerical study of the power function of a new symmetry test'. Together they form a unique fingerprint.

Cite this