E-values : Calibration, combination, and applications. / Vovk, Vladimir; Wang, Ruodu.

In: The Annals of Statistics, Vol. 49, No. 3, 09.08.2021, p. 1736-1754.

Research output: Contribution to journalArticlepeer-review

Published

Standard

E-values : Calibration, combination, and applications. / Vovk, Vladimir; Wang, Ruodu.

In: The Annals of Statistics, Vol. 49, No. 3, 09.08.2021, p. 1736-1754.

Research output: Contribution to journalArticlepeer-review

Harvard

Vovk, V & Wang, R 2021, 'E-values: Calibration, combination, and applications', The Annals of Statistics, vol. 49, no. 3, pp. 1736-1754. https://doi.org/10.1214/20-AOS2020

APA

Vovk, V., & Wang, R. (2021). E-values: Calibration, combination, and applications. The Annals of Statistics, 49(3), 1736-1754. https://doi.org/10.1214/20-AOS2020

Vancouver

Vovk V, Wang R. E-values: Calibration, combination, and applications. The Annals of Statistics. 2021 Aug 9;49(3):1736-1754. https://doi.org/10.1214/20-AOS2020

Author

Vovk, Vladimir ; Wang, Ruodu. / E-values : Calibration, combination, and applications. In: The Annals of Statistics. 2021 ; Vol. 49, No. 3. pp. 1736-1754.

BibTeX

@article{f36284dca6394713832d62a3157ac569,
title = "E-values: Calibration, combination, and applications",
abstract = "Multiple testing of a single hypothesis and testing multiple hypotheses are usually done in terms of p-values. In this paper we replace p-values with their natural competitor, e-values, which are closely related to betting, Bayes factors, and likelihood ratios. We demonstrate that e-values are often mathematically more tractable; in particular, in multiple testing of a single hypothesis, e-values can be merged simply by averaging them. This allows us to develop efficient procedures using e-values for testing multiple hypotheses.",
keywords = "hypothesis testing, multiple hypothesis testing, global null, admissible decisions, test martingale, Bayes factor",
author = "Vladimir Vovk and Ruodu Wang",
year = "2021",
month = aug,
day = "9",
doi = "10.1214/20-AOS2020",
language = "English",
volume = "49",
pages = "1736--1754",
journal = "The Annals of Statistics",
issn = "0090-5364",
publisher = "Institute of Mathematical Statistics",
number = "3",

}

RIS

TY - JOUR

T1 - E-values

T2 - Calibration, combination, and applications

AU - Vovk, Vladimir

AU - Wang, Ruodu

PY - 2021/8/9

Y1 - 2021/8/9

N2 - Multiple testing of a single hypothesis and testing multiple hypotheses are usually done in terms of p-values. In this paper we replace p-values with their natural competitor, e-values, which are closely related to betting, Bayes factors, and likelihood ratios. We demonstrate that e-values are often mathematically more tractable; in particular, in multiple testing of a single hypothesis, e-values can be merged simply by averaging them. This allows us to develop efficient procedures using e-values for testing multiple hypotheses.

AB - Multiple testing of a single hypothesis and testing multiple hypotheses are usually done in terms of p-values. In this paper we replace p-values with their natural competitor, e-values, which are closely related to betting, Bayes factors, and likelihood ratios. We demonstrate that e-values are often mathematically more tractable; in particular, in multiple testing of a single hypothesis, e-values can be merged simply by averaging them. This allows us to develop efficient procedures using e-values for testing multiple hypotheses.

KW - hypothesis testing

KW - multiple hypothesis testing

KW - global null

KW - admissible decisions

KW - test martingale

KW - Bayes factor

U2 - 10.1214/20-AOS2020

DO - 10.1214/20-AOS2020

M3 - Article

VL - 49

SP - 1736

EP - 1754

JO - The Annals of Statistics

JF - The Annals of Statistics

SN - 0090-5364

IS - 3

ER -