Admissible ways of merging p-values under arbitrary dependence
Authors:
- 1 February 2022
- journal article
- research article
- Published by Institute of Mathematical Statistics in The Annals of Statistics
- Vol. 50 (1) , 351-375
Abstract
Methods of merging several p-values into a single p-value are important in their own right and widely used in multiple hypothesis testing. This paper is the first to systematically study the admissibility (in Wald’s sense) of p-merging functions and their domination structure, without any information on the dependence structure of the input p-values. As a technical tool, we use the notion of e-values, which are alternatives to p-values recently promoted by several authors. We obtain several results on the representation of admissible p-merging functions via e-values and on (in)admissibility of existing p-merging functions. By introducing new admissible p-merging functions, we show that some classic merging methods can be strictly improved to enhance power without compromising validity under arbitrary dependence.Keywords
All Related Versions
This publication has 23 references indexed in Scilit:
- Joint MixabilityMathematics of Operations Research, 2016
- General convex order on risk aggregationScandinavian Actuarial Journal, 2015
- Multiple Testing for Exploratory ResearchStatistical Science, 2011
- The complete mixability and convex minimization problems with monotone marginal densitiesJournal of Multivariate Analysis, 2011
- On the distributional transform, Sklar's theorem, and the empirical copula processJournal of Statistical Planning and Inference, 2009
- A stochastic process approach to false discovery controlThe Annals of Statistics, 2004
- Some probability inequalities for ordered $\rm MTP\sb 2$ random variables: a proof of the Simes conjectureThe Annals of Statistics, 1998
- An improved Bonferroni procedure for multiple tests of significanceBiometrika, 1986
- Das maximale signifikanzniveau des Tests: “LehneH o ab, wennk untern gegebenen tests zur ablehnung führen”Metrika, 1978
- Significance Tests in Parallel and in SeriesJournal of the American Statistical Association, 1958