scientific article; zbMATH DE number 1406070

From MaRDI portal
Publication:4939084





Fom the introduction: Rank-based nonparametric procedures have served statisticians well, providing easy-to-use, intuitive, efficient, and robust alternatives to the common normal theory procedures. The original rank procedures were univariate. Multivariate procedures, in which each variable is ranked separately, immediately affords practitioners many of the benefits of the univariate procedures in more complicated situations. In recent years, there has been development of other methods of multivariate ranking that use the variables together rather than separately. This paper shows how to use particular definitions of multivariate sign and rank to generate a multitude of multivariate procedures that not only mimic the popular univariate ones, but go beyond to treat genuinely multivariate problems. For the entire collection see [Zbl 0927.00053].




Cited in
(25)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4939084)