Exact and Approximate Stepdown Methods for Multiple Hypothesis Testing
多重假设检验的精确与近似逐步下降方法AI 解读 ↓
← 返回论文列表与筛选
阅读说明
本页整理公开书目信息，不生成学术质量评分。AI 解读根据题名及所注明的摘要或原文片段生成，未完成全文核验，可能有误；具体假设、证明与结论请核对原文。摘要保留来源文本，未经模型改写。
DOI 首次登记不等于接收或发表日期。部分更正、评论等条目没有独立摘要；缺少可靠材料时保留待补状态。
引用量采用 OpenAlex 单条关联记录，不合计多个记录；关联记录可能包含正式发表版本，不等于此版本单独引用。缺失不视为零，引用指标不代表论文质量。
关注作者按书目中的完整姓名匹配，同名和姓名变体需人工辨别。相关论文先按共享细分方向和具体方法排列，再参考通用方法与发表日期。
分类依据与纠错
AI 根据所提供的论文材料归类：摘要提供通用多重检验与重抽样理论，没有指定因果、面板、时间序列或金融问题，故不强行归类。
反馈分类问题
正在核对关联版本…
代码、数据与补充材料
正在核对对应版本的资源…
关注这篇论文的作者
多重假设检验的精确与近似逐步下降方法
- 研究问题
- 同时检验多个假设时，如何在不要求子集枢轴性的情况下构造逐步下降检验，并控制族错误率及计算负担？
- 方法与思路
- 作者对估计临界值施加单调性，逐阶段构造控制第一类错误的交集检验，并允许各阶段复用同一组重抽样或子抽样。
- 材料陈述的贡献
- 摘要称，所得一般逐步下降方法不依赖子集枢轴性，可分阶段实施且至多计算与假设数量相同次数的检验。
- 阅读关注点
- 应核对族错误率控制的有限样本与大样本条件、临界值单调化规则，以及交集检验和重抽样的有效性要求。
- 阅读边界
- 仅据版本未注明的 OpenAlex 摘要；不要求子集枢轴性并不意味着无其他条件，也不能据此断言功效普遍更高。
原文摘要
Consider the problem of testing k hypotheses simultaneously. In this article we discuss finite- and large-sample theory of stepdown methods that provide control of the familywise error rate (FWE). To improve on the Bonferroni method or on Holm's stepdown method, Westfall and Young made effective use of resampling to construct stepdown methods that implicitly estimate the dependence structure of the test statistics. However, their methods depend on an assumption known as “subset pivotality.” Our goal here is to construct general stepdown methods that do not require such an assumption. To accomplish this, we take a close look at what makes stepdown procedures work; a key component is a monotonicity requirement of critical values. By imposing monotonicity on estimated critical values (which is not an assumption on the model but rather is an assumption on the method), we show how to construct stepdown tests that can be applied in a stagewise fashion so that at most k tests need to be computed. Moreover, at each stage, an intersection test that controls the usual probability of a type 1 error is calculated, which allows us to draw on an enormous resampling literature as a general means of test construction. In addition, it is possible to carry out this method using the same set of resamples (or subsamples) for each of the intersection tests.