Title
Stepwise multiple tests for successive comparisons of treatment effects
Abbreviated Journal Title
Comput. Stat. Data Anal.
Keywords
critical constants; familywise error rate; multiple comparisons; multivariate-t distribution; simultaneous confidence intervals; Computer Science, Interdisciplinary Applications; Statistics &; Probability
Abstract
Suppose that the k treatments under comparison are ordered in a certain way. For example, they may be a sequence of increasing dose levels of a drug. It is interesting to look directly at the successive differences between the treatment effects mu(i)'s, namely the set of differences mu(2) - mu(1), mu(3) - mu(2)...,mu(k) - mu(k-1). Lee and Spurrier (J, Statist. Plann. Inference 43 (1995) 323) propose a one-sided and a two-sided simultaneous confidence interval procedures for making successive comparisons between treatments. In this paper we develop step-down and step-up tests for testing the families of hypotheses H-i0: mu(i+1) - mu(i) = 0 vs H-ia: mu(i+1) > 0, i = 1,...,k - I (one-sided alternatives) H-i0: mu(i+1) - mu(i) = 0 vs H-ia: mu(i+1) - mu(i) not equal 0, i = 1,...,k - 1 (two-sided alternatives) These stepwise tests are uniformly more powerful than the simultaneous confidence interval procedures in terms of rejection of the null hypotheses, though the simultaneous confidence interval procedures provide information on the magnitudes of the mu(i+1) - mu(i)'s. The critical constants required for applying these multiple tests are provided, and the tests are illustrated with a numerical example. (C) 2003 Published by Elsevier B.V.
Journal Title
Computational Statistics & Data Analysis
Volume
46
Issue/Number
1
Publication Date
1-1-2004
Document Type
Article
Language
English
First Page
189
Last Page
199
WOS Identifier
ISSN
0167-9473
Recommended Citation
"Stepwise multiple tests for successive comparisons of treatment effects" (2004). Faculty Bibliography 2000s. 4554.
https://stars.library.ucf.edu/facultybib2000/4554
Comments
Authors: contact us about adding a copy of your work at STARS@ucf.edu