Merging sequential e-values via martingales

被引:1
作者
Vovk, Vladimir [1 ]
Wang, Ruodu [2 ]
机构
[1] Royal Holloway Univ London, Dept Comp Sci, Egham, Surrey, England
[2] Univ Waterloo, Dept Stat & Actuarial Sci, Waterloo, ON, Canada
来源
ELECTRONIC JOURNAL OF STATISTICS | 2024年 / 18卷 / 01期
基金
加拿大自然科学与工程研究理事会;
关键词
Anytime validity; betting scores; e-processes; admissibility; merging functions;
D O I
10.1214/24-EJS2228
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
We study the problem of merging sequential or independent evalues into one e -value or e -process. We describe a class of e -value merging functions via martingales and show that it dominates all merging methods for sequential e -values. All admissible methods for constructing e -processes can also be obtained in this way. In the case of merging independent evalues, the situation becomes much more complicated, and we provide a general class of such merging functions based on martingales applied to reordered data.
引用
收藏
页码:1185 / 1205
页数:21
相关论文
共 21 条
[1]  
GRUNWALD P., 2023, Journal of the Royal Statistical Society
[2]   Valid sequential inference on probability forecast performance [J].
Henzi, Alexander ;
Ziegel, Johanna F. .
BIOMETRIKA, 2022, 109 (03) :647-663
[3]   TIME-UNIFORM, NONPARAMETRIC, NONASYMPTOTIC CONFIDENCE SEQUENCES [J].
Howard, Steven R. ;
Ramdas, Aaditya ;
McAuliffe, Jon ;
Sekhon, Jasjeet .
ANNALS OF STATISTICS, 2021, 49 (02) :1055-1080
[4]   E-values as unnormalized weights in multiple testing [J].
Ignatiadis, Nikolaos ;
Wang, Ruodu ;
Ramdas, Aaditya .
BIOMETRIKA, 2024, 111 (02) :417-439
[5]  
Kolmogorov A.N, 1963, The Indian Journal of Statistics, Series A, P369
[6]   A NEW INTERPRETATION OF VON MISES CONCEPT OF RANDOM SEQUENCE [J].
LOVELAND, D .
ZEITSCHRIFT FUR MATHEMATISCHE LOGIK UND GRUNDLAGEN DER MATHEMATIK, 1966, 12 (04) :279-&
[8]  
Ramdas A., 2022, ARXIV
[9]   Derandomised knockoffs: leveraging e-values for false discovery rate control [J].
Ren, Zhimei ;
Barber, Rina Foygel .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 2024, 86 (01) :122-154
[10]  
Rogers L. C. G., 2000, DIFFUSIONS MARKOV PR