Einfache Begründung für die Irrtumswahrscheinlichkeitsschranke der Hommel'schen Mehrheitstestkombination. (A simple justification for the error probability bound of Hommel's majority test combination)

DOI10.1007/BF01897810MaRDI QIDQ1822159zbMATH OpenFDO

Authors Dietrich Morgenstern

Publication date 1985

Published in Metrika (Search for Journal in Brave)

Full work available at URL https://eudml.org/doc/176000

zbMATH Keywords

majority-tests

Mathematics Subject Classification ID

Compound decision problems in statistical decision theory (62C25) Parametric hypothesis testing (62F03)

Review text

Starting from n different tests, \textit{G. Hommel} [Entwicklung einer statistischen Teststrategie für komplexe Fragestellungen. Habil.- Schrift, Univ. Erlangen/Nürnberg (1979)] combined the different majority-tests [\textit{B. Rüger}: reject if at least k of the n tests reject, Metrika 25, 171-178 (1978; Zbl 0389.62013)] with the same level of significance to a new test whose significance level he majorized. Here a very simple proof for his result is given.

Recommendations

• scientific article; zbMATH DE number 1366702
• An improved Hochberg procedure for multiple tests of significance
• A sharper Bonferroni procedure for multiple tests of significance
• On actual significance levels of combined tests
• Control of directional errors with stagewise multiple test procedures

Cites work

• Berechnung des maximalen Signifikanzniveaus des Testes Lehne \(H_0\) ab, wenn k unter n gegebenen Tests zur Ablehnung führen
• Das maximale Signifikanzniveau des Tests: Lehne \(H_0\) ab, wenn \(k\) unter \(n\) gegebenen Tests zur Ablehnung führen

This page was built for publication: Einfache Begründung für die Irrtumswahrscheinlichkeitsschranke der Hommel'schen Mehrheitstestkombination. (A simple justification for the error probability bound of Hommel's majority test combination)

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1822159)