Game-theoretic statistics and safe anytime-valid inference
From MaRDI portal
Publication:6145149
Abstract: Safe anytime-valid inference (SAVI) provides measures of statistical evidence and certainty -- e-processes for testing and confidence sequences for estimation -- that remain valid at all stopping times, accommodating continuous monitoring and analysis of accumulating data and optional stopping or continuation for any reason. These measures crucially rely on test martingales, which are nonnegative martingales starting at one. Since a test martingale is the wealth process of a player in a betting game, SAVI centrally employs game-theoretic intuition, language and mathematics. We summarize the SAVI goals and philosophy, and report recent advances in testing composite hypotheses and estimating functionals in nonparametric settings.
Cites work
- scientific article; zbMATH DE number 3179081 (Why is no real title available?)
- scientific article; zbMATH DE number 720689 (Why is no real title available?)
- scientific article; zbMATH DE number 791368 (Why is no real title available?)
- scientific article; zbMATH DE number 7626761 (Why is no real title available?)
- scientific article; zbMATH DE number 3189754 (Why is no real title available?)
- scientific article; zbMATH DE number 3034028 (Why is no real title available?)
- scientific article; zbMATH DE number 3045589 (Why is no real title available?)
- scientific article; zbMATH DE number 3072532 (Why is no real title available?)
- A Sequential Non-Parametric Multivariate Two-Sample Test
- A TCHEBYCHEFF-LIKE INEQUALITY FOR STOCHASTIC PROCESSES
- A general class of exponential inequalities for martingales and ratios
- Algorithmic Learning in a Random World
- Always Valid Inference: Continuous Monitoring of A/B Tests
- Bayes factors and marginal distributions in invariant situations
- CONFIDENCE SEQUENCES FOR MEAN, VARIANCE, AND MEDIAN
- Catoni-style confidence sequences for heavy-tailed mean estimation
- Challenging the empirical mean and empirical variance: a deviation study
- E-values: calibration, combination and applications
- Exact anytime-valid confidence intervals for contingency tables and beyond
- Exponential inequalities for martingales with applications
- Exponential inequalities for sums of weakly dependent variables
- False Discovery Rate–Adjusted Multiple Confidence Intervals for Selected Parameters
- Fixed-Sample-Size Analysis of Sequential Observations
- Game-theoretic foundations for probability and finance
- Gaussian universal likelihood ratio testing
- Insuring against loss of evidence in game-theoretic probability
- Interactive martingale tests for the global null
- Likelihood, Replicability and Robbins' Confidence Sequences
- ON A SEQUENTIAL t-TEST
- On confidence sequences
- Panning for Gold: ‘Model-X’ Knockoffs for High Dimensional Controlled Variable Selection
- Present Position and Potential Developments: Some Personal Views: Statistical Theory: The Prequential Approach
- Probability and finance. It's only a game!
- SOME NONPARAMETRIC SEQUENTIAL TESTS WITH POWER ONE
- Sequential Tests of Statistical Hypotheses
- Sequential estimation of quantiles with applications to A/B testing and best-arm identification
- Some aspects of the sequential design of experiments
- Statistical Methods Related to the Law of the Iterated Logarithm
- Student's t-Test Under Symmetry Conditions
- Test martingales, Bayes factors and \(p\)-values
- Testing exchangeability: fork-convexity, supermartingales and e-processes
- Testing randomness online
- The control of the false discovery rate in multiple testing under dependency.
- The expected sample size of some tests of power one
- The minimum description length principle in coding and modeling
- Time-uniform Chernoff bounds via nonnegative supermartingales
- Time-uniform, nonparametric, nonasymptotic confidence sequences
- Universal coding, information, prediction, and estimation
- Universal inference
- Valid sequential inference on probability forecast performance
Cited in
(3)
This page was built for publication: Game-theoretic statistics and safe anytime-valid inference
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6145149)
