Scholion.ai

A new permutation test statistic for complete block designs

Inga Samonenko, John Robinson

The Annals of Statistics · 2015 · Volume 43 · Issue 1

DOI: 10.1214/14-AOS1266

Abstract

We introduce a nonparametric test statistic for the permutation test in complete block designs. We find the region in which the statistic exists and consider particularly its properties on the boundary of the region. Further, we prove that saddlepoint approximations for tail probabilities can be obtained inside the interior of this region. Finally, numerical examples are given showing that both accuracy and power of the new statistic improves on these properties of the classical $F$-statistic under some non-Gaussian models and equals them for the Gaussian case.

Keywords

60F10, 62G09, 62G10, 62G20, admissible domain, saddlepoint approximations

Enable JavaScript for citation tools and additional paper interactions.