Testing with p^*-values: between p-values, mid p-values, and e-values
From MaRDI portal
Publication:6201852
DOI10.3150/23-BEJ1633arXiv2010.14010OpenAlexW4391458257MaRDI QIDQ6201852FDOQ6201852
Authors: Ruodu Wang 
Publication date: 26 March 2024
Published in: Bernoulli (Search for Journal in Brave)
Abstract: We introduce the notion of p*-values (p*-variables), which generalizes p-values (p-variables) in several senses. The new notion has four natural interpretations: operational, probabilistic, Bayesian, and frequentist. A main example of a p*-value is a mid p-value, which arises in the presence of discrete test statistics. A unified stochastic representation for p-values, mid p-values, and p*-values is obtained to illustrate the relationship between the three objects. We study several ways of merging arbitrarily dependent or independent p*-values into one p-value or p*-value. Admissible calibrators of p*-values to and from p-values and e-values are obtained with nice mathematical forms, revealing the role of p*-values as a bridge between p-values and e-values. The notion of p*-values becomes useful in many situations even if one is only interested in p-values, mid p-values, or e-values. In particular, deterministic tests based on p*-values can be applied to improve some classic methods for p-values and e-values.
Full work available at URL: https://arxiv.org/abs/2010.14010
Parametric hypothesis testing (62F03) Bayesian inference (62F15) Hypothesis testing in multivariate analysis (62H15) Inequalities; stochastic orderings (60E15) Paired and multiple comparisons; multiple testing (62J15)
Cites Work
- New FDR bounds for discrete and heterogeneous tests
- An improved Bonferroni procedure for multiple tests of significance
- Posterior predictive \(p\)-values
- The control of the false discovery rate in multiple testing under dependency.
- A stochastic process approach to false discovery control.
- Time-uniform, nonparametric, nonasymptotic confidence sequences
- Multiple Hypotheses Testing with Weights
- Test martingales, Bayes factors and \(p\)-values
- E-values: calibration, combination and applications
- Random variables with maximum sums
- Halving the Bounds for the Markov, Chebyshev, and Chernoff Inequalities Using Smoothing
- Combining p-values via averaging
- Multiple test functions and adjusted \(p\)-values for test statistics with discrete distributions
- Large-scale inference. Empirical Bayes methods for estimation, testing, and prediction
- Comparison methods for stochastic models and risks
- Stochastic orders
- The Existence of Probability Measures with Given Marginals
- Mathematical Risk Analysis
- Some probability inequalities for ordered MTP\(_2\) random variables: a proof of the Simes conjecture
- Sum of arbitrarily dependent random variables
- Interactive martingale tests for the global null
- Multiple testing for exploratory research
- The harmonic mean p -value for combining dependent tests
- Algorithmic Learning in a Random World
- Sums of standard uniform random variables
- Testing for outliers with conformal p-values
- Confidence and discoveries with \(e\)-values
- A unified treatment of multiple testing with prior knowledge using the p-filter
- Testing randomness online
- Universal inference
- Cauchy Combination Test: A Powerful Test With Analytic p-Value Calculation Under Arbitrary Dependency Structures
- A Theory for Measures of Tail Risk
- Admissible ways of merging \(p\)-values under arbitrary dependence
- Trade-off Between Validity and Efficiency of Merging P-Values Under Arbitrary Dependence
- Meta-Analysis of Mid-p-Values: Some New Results based on the Convex Order
Cited In (1)
This page was built for publication: Testing with \(\mathrm{p}^*\)-values: between p-values, mid p-values, and e-values
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6201852)
