The Benjamini-Hochberg method with infinitely many contrasts in linear models

• 成果类型: Article
• 署名作者: Westfall, Peter H.
• 署名单位: Texas Tech University System; Texas Tech University
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/asn033
• 发表日期: 2008
• 页码: 709719
• 关键词: false discovery rate; multiple comparisons; familywise; error

摘要

Benjamini and Hochberg's method for controlling the false discovery rate is applied to the problem of testing infinitely many contrasts in linear models. Exact, easily calculated critical values are derived, defining a new multiple comparisons method for testing contrasts in linear models. The method is adaptive, depending on the data through the F-statistic, like the Waller-Duncan Bayesian multiple comparisons method. Comparisons with Scheffe's method are given, and the method is extended to the simultaneous confidence intervals of Benjamini and Yekutieli.