Stepwise multiple testing procedures have attracted several statisticians for decades and are also quite popular with statistics users because of their technical simplicity. The Bonferroni procedure has been one of the earliest and most prominent testing rules for controlling the familywise error rate (FWER). A recent article established that the FWER for the Bonferroni method asymptotically (i.e., when the number of hypotheses becomes arbitrarily large) approaches zero under any positively equicorrelated multivariate normal framework. However, similar results for the limiting behaviors of FWER of general stepwise procedures are nonexistent. The present work addresses this gap in a unified manner by studying the limiting behaviors of the FWER of several stepwise testing rules for correlated normal setups. Specifically, we show that the limiting FWER approaches zero for any step-down rule (e.g., Holm's method) provided the infimum of the correlations is strictly positive. We also establish similar limiting zero results on FWER of other popular multiple testing rules, e.g., Hochberg's and Hommel's procedures. We then extend these results to any configuration of true and false null hypotheses. It turns out that, within our chosen asymptotic framework, the Benjamini-Hochberg method can hold the FWER at a strictly positive level asymptotically under the equicorrelated normality. We finally discuss the limiting powers of various procedures.
翻译:几十年来,渐进式多重测试程序吸引了数名统计学家,并且由于其技术简单性,也非常受统计用户的欢迎。 Bonferroni 程序是控制家庭误差率的最早和最突出的测试规则之一。最近一篇文章规定,对于Bonferroni 方法的FWER在任何正等值相关多变正常框架下(即假设数量随任意而大)接近零度(即当假设数量随时间而增加时),在任何正等值相关多变正常框架下(即,假设数量随时间而异)接近零度。然而,对于FWER的一般渐进式程序限制行为,目前没有类似的类似结果。目前的工作通过研究FWER对相关正常配置的几条渐进式测试规则的限制性行为,以统一的方式解决这一差距。具体地说,我们表明,FWER的限值方法在任何递减规则(例如Holmums)中都接近零度,我们也可以在其它流行的多重测试规则的FWER上设定类似的限制零度。例如,Amberberg和Homlimal-Hal-chemal 程序最后将这些结果扩展到任何正确的格式。我们所选择的Rest-chal-chal-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx