Answers are grounded in this paper and cite their sources.
Let (X⁽¹⁾ⱼ₁, ⋯, X⁽ᵖ⁾ⱼ₁), ⋯, (X⁽¹⁾jnⱼ, ⋯, X⁽ᵖ⁾jnⱼ) be random samples of size nⱼ from the c continuous p-variate distribution functions Fⱼ(x) = F(x - θⱼ) where x = (x₁, ⋯, xₚ), θⱼ = (θᵢⱼ, ⋯, θₚⱼ) and j = 1, ⋯, c. This paper is concerned with the problem of testing the hypothesis H : θ₁ = θ₂ = ⋯ = θc against the alternative that all θⱼ are not equal on the basis of the above c samples. When we especially consider the asymptotic efficiency of test, the following alternative K is adopted, K : θⱼ = θ + νⱼ/N1/2,νⱼ = (νᵢⱼ, ⋯, νₚⱼ), j = 1, ⋯, c. We shall develop in this paper some test procedures for the hypothesis H which are originated from the paper of Chernoff-Savage [2]. When $c = 2$, that is multivariate two-sample tests, Sugiura [7] has proposed a test statistic of Wilcoxon type. On the other hand, Puri [5] has derived univariate several-sample tests in general type including the test of Kruskal-Wallis [4]. It will be shown that the latter are corresponding to the case $p = 1$ in this paper and the former is a particular one among the case $c = 2$ of our tests.
No takes yet. Share an insight, caveat, or question.
Ryoji Tamura (1966) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: