ON THE MODIFIED S-METHOD AND DIRECTIONAL ERRORS

被引：10

作者：

FINNER, H

机构：

[1] University of Trier, Fachbereich IV - Mathematik/Statistik

来源：

COMMUNICATIONS IN STATISTICS-THEORY AND METHODS | 1990年 / 19卷 / 01期

关键词：

Closed multiple test procedure; coherence; consonance; LSD-test; multiple level a; one-sided hypothesis; range-test;

D O I：

10.1080/03610929008830186

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

Scheffe (1970) introduced a modified S-method for testing all linear contrasts, which is a stagewise rejective two stage test procedure. If a contrast hypothesis a' μ = 0 is rejected using this method, the problem arises whether or not one can decide for a’ p < 0 or aμ>0 without violating the multiple level a. We give a general proof that the modified S-method still controls the experimentwise level of significance a. Furthermore, we show that this pro. © 1990, Taylor & Francis Group, LLC. All rights reserved.

引用

页码：41 / 53

页数：13

共 14 条

[1]

Bauer P., 1986, METRIKA, V33, P121

[2] MULTIPLE RANGE AND MULTIPLE F TESTS [J].