A new rank test for the K-Sample problem (Q3702286)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3928099
Language Label Description Also known as
default for all languages
No label defined
    English
    A new rank test for the K-Sample problem
    scientific article; zbMATH DE number 3928099

      Statements

      A new rank test for the K-Sample problem (English)
      0 references
      0 references
      0 references
      0 references
      0 references
      1985
      0 references
      asymptotic normality
      0 references
      dissociated random variables
      0 references
      asymptotic
      0 references
      chi- square
      0 references
      treatment sum of squares
      0 references
      one-way-analysis of variance
      0 references
      generalization of the two-sample Wilcoxon rank sum statistic
      0 references
      k-sample test
      0 references
      simulation results
      0 references
      power comparisons
      0 references
      examples
      0 references
      In this paper a particular representation of the treatment sum of squares in the one-way-analysis of variance is used to suggest a generalization of the two-sample Wilcoxon rank sum statistic to provide a k-sample test different from the Kruskal-Wallis and from the Jonkheere-test. Its test statistic is \[ D=12 N^{-1}\sum_{1\leq i<j\leq k}(n_ i+n_ j)(W(i,j)-2^{-1}n_ in_ j)^ 2/n\quad_ in_ j(n_ i+n_ j+1), \] where W(i,j) is the Wilcoxon statistic for samples i and j. D is shown to be asymptotically \(\chi^ 2\)-distributed when \(n_ i\to \infty\) and asymptotically normal, as \(k\to \infty.\) Some simulation results concerning power comparisons and some examples are given, too.
      0 references

      Identifiers