ON THE POWER OF A NEW STATISTICAL TEST AND TWO-SAMPLE WILCOXON TEST
G. I. Salov
a:2:{s:4:"TEXT";s:169:"Institute of Computational Mathematics and Mathematical Geophysics, Siberian Branch, Russian Academy of Sciences, pr. Akademika Lavrent’eva 6, Novosibirsk, 630090 Russia";s:4:"TYPE";s:4:"text";}
Keywords: two samples, three samples, nonparametric tests, Wilcoxon test, new statistical test
Subsection: ANALYSIS AND SYNTHESIS OF SIGNALS AND IMAGES
Abstract
One approach to finding suitable critical values of statistics of a new nonparametric statistical test previously proposed by the author for the problem of testing the hypothesis of homogeneity of two and three samples is considered. The Wilcoxon test is equivalent to a particular ("linear") case of the proposed test. A comparative analysis of these tests is performed for the case where the Wilcoxon test is locally the most powerful one among the ranking methods.
|
