Inverse sampling for multivariate ninparametric two-sample problems

From MaRDI portal





Let \(F_ 1\) and \(F_ 2\) be two multivariate d.f.'s. The problem of testing the null hypothesis \(H: F_ 1=F_ 2\) against the alternative \(K: F_ 1\neq F_ 2\) is considered. A nonparametric test criterion of sequential type, when there is a sample of fixed size from the d.f. \(F_ 1\) and the observations from the d.f. \(F_ 2\) are drawn sequentially is suggested. The test is a multivariate extension of that studied by \textit{S. K. Chatterjee} and the author [Bull., Calcutta Stat. Assoc. 33, 35-58 (1984; Zbl 0561.62072)]. Asymptotic distributions of the proposed test statistics both under the null hypothesis H and under fixed and contiguous alternatives are obtained.











This page was built for publication: Inverse sampling for multivariate ninparametric two-sample problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3723524)