Beyond Neyman-Pearson: e-values enable hypothesis testing with a data-driven alpha

DOI10.1073/PNAS.2302098121MaRDI QIDQ6844959zbMATH OpenFDO

Authors Peter D. Grünwald

Publication date 13 March 2026

Published in Proceedings of the National Academy of Sciences of the United States of America (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1073/PNAS.2302098121

Mathematics Subject Classification ID

Parametric hypothesis testing (62F03)

Cited in

(8)

• Choosing alpha post hoc: the danger of multiple standard significance thresholds
• On the optimality of coin-betting for mean estimation
• The only admissible way of merging arbitrary e-values
• Bayesian Model Checking by Betting: A Game-Theoretic Alternative to Bayesian p -values and Classical Bayes Factors
• On admissibility in post-hoc hypothesis testing
• Measuring evidence against exchangeability and group invariance with e-values
• Discovery on the basis of the established amount of information in the evidence against the null hypothesis: optimal fact-finding in the presence of imprecise probability
• Carefree multiple testing with e-processes

This page was built for publication: Beyond Neyman-Pearson: e-values enable hypothesis testing with a data-driven alpha

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6844959)