论文信息 - Enhancement of prediction algorithms by betting

Enhancement of prediction algorithms by betting

This note proposes a procedure for enhancing the quality of probabilistic prediction algorithms via betting against their predictions. It is inspired by the success of the conformal test martingales that have been developed recently. The version of this note at http://alrw.net (Working Paper 34) is updated most often.

Vladimir Vovk

[1] Vladimir Vovk,et al. Testing Randomness Online , 2019, Statistical Science.

[2] Vladimir Vovk. Conformal testing in a binary model situation , 2021, ArXiv.

[3] Vladimir Vovk,et al. Game‐Theoretic Foundations for Probability and Finance , 2019, Wiley Series in Probability and Statistics.

[4] A. N. Kolmogorov,et al. Foundations of the theory of probability , 1960 .

[5] Helly. Grundbegriffe der Wahrscheinlichkeitsrechnung , 1936 .

[6] A. Dawid,et al. Prequential probability: principles and properties , 1999 .

[7] W. Gasarch,et al. The Book Review Column 1 Coverage Untyped Systems Simple Types Recursive Types Higher-order Systems General Impression 3 Organization, and Contents of the Book , 2022 .

[8] M. Rosenblatt. Remarks on a Multivariate Transformation , 1952 .

[9] A. Raftery,et al. Strictly Proper Scoring Rules, Prediction, and Estimation , 2007 .

[10] Vladimir Vovk,et al. Testing for concept shift online , 2020, ArXiv.

[11] P. Levy. Théorie de l'addition des variables aléatoires , 1955 .

[12] Alexander Gammerman,et al. Retrain or not retrain: Conformal test martingales for change-point detection , 2021, COPA.