A new permutation test statistic for complete block designs. (English) Zbl 1308.62156

Summary: 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.


62K10 Statistical block designs
62G10 Nonparametric hypothesis testing
62G20 Asymptotic properties of nonparametric inference


Full Text: DOI arXiv Euclid


[1] Borovkov, A. A. and Rogozin, B. A. (1965). On the multi-dimensional central limit theorem. Theory Probab. Appl. 10 55-62. · Zbl 0139.35206
[2] Brown, B. M. and Maritz, J. S. (1982). Distribution-free methods in regression. Austral. J. Statist. 24 318-331. · Zbl 0507.62038
[3] Fisher, R. A. (1935). The Design of Experiments . Oliver and Boyd, Edinburgh.
[4] Genz, A. (2003). Fully symmetric interpolatory rules for multiple integrals over hyper-spherical surfaces. J. Comput. Appl. Math. 157 187-195. · Zbl 1028.65015
[5] Jin, R. and Robinson, J. (1999). Saddlepoint approximation near the endpoints of the support. Statist. Probab. Lett. 45 295-303. · Zbl 0951.62012
[6] Kolassa, J. and Robinson, J. (2011). Saddlepoint approximations for likelihood ratio like statistics with applications to permutation tests. Ann. Statist. 39 3357-3368. · Zbl 1246.62121
[7] Robinson, J. (1982). Saddlepoint approximations for permutation tests and confidence intervals. J. R. Stat. Soc. Ser. B Stat. Methodol. 44 91-101. · Zbl 0487.62016
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.