On the Powers of Some New Chi-Square Type Statistics

MaRDI QIDQ3605910zbMATH OpenFDO

Authors Clement Ampadu

Publication date 25 February 2009

Full work available at URL https://arxiv.org/abs/1007.5107, http://pphmj.com/abstract/3456.htm

zbMATH Keywords

alternative distributionchi-square-type statisticsgoodness-of-fitmultinomial distributionnull distributionpower

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10)

Abstract: In this paper, four new Chi-Square type statistics are presented for testing the hypothesis of a uniform null versus specified trend alternatives. The powers of these test statistics are compared with the powers of the statistics considered by Steele and Chaseling [8]. The four test statistics are shown to have superior or equivalent powers to the powers of the test statistics considered by the authors for certain trend alternatives and for certain conditions placed on the cell count.

Recommendations

• A new goodness of fit test for the equality of multinomial cell probabilities versus trend alternatives
• Freeman-Tukey chi-squared goodness-of-fit statistics
• scientific article; zbMATH DE number 5518683
• Powers of Discrete Goodness-of-Fit Test Statistics for a Uniform Null Against a Selection of Alternative Distributions
• On the Chi-square test of homogeneity in case of a simultaneous increase of the number of observations and the number of interval partitions

Cited in

(3)

• A new goodness of fit test for the equality of multinomial cell probabilities versus trend alternatives
• Simulated power of some discrete goodness-of-fit test statistics for testing the null hypothesis of a `zig-zag' distribution
• On the difference of two chi-square variates with application to outage probability computation

This page was built for publication: On the Powers of Some New Chi-Square Type Statistics

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3605910)