figshare
Browse
Table 3.xls (13.5 kB)

Comparison of the powers of the tests for normality for GG distribution and N = 50.

Download (13.5 kB)
dataset
posted on 2020-06-11, 20:06 authored by Agnieszka Wyłomańska, D. Robert Iskander, Krzysztof Burnecki

The following tests are taken under consideration: the new test proposed in this paper, Jarque-Bera (JB) test [5], D’Agostino-Pearson (DP) test [4], Shapiro-Wilk (SW) test [3], test based on the empirical characteristic function (CF) [22], Kolmogorov-Smirnov (KS) test [2], Kuiper test [6], Watson test [7], Cramer-von Mises (CvM) test [8], Anderson-Darling (AD) test [9] and χ2 goodness-of-fit test [1].

History

Usage metrics

    PLOS ONE

    Categories

    Keywords

    hypothesis teststest statisticprobability distributionnovel approachsignal processing techniquesdata examplesleptokurtic onesplatykurtic distributionstesting normalityEdgeworth expansion Statistical inferenceOmnibus testEdgeworth expansionsample sizesconfidence intervalstime seriesleptokurtic distributions

    Licence

    CC BY 4.0

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC