Catoni-style confidence sequences for heavy-tailed mean estimation

DOI10.1016/j.spa.2023.05.007arXiv2202.01250OpenAlexW4380149598MaRDI QIDQ6171648

Hong-Jian Wang, Aaditya Ramdas

Publication date: 14 August 2023

Published in: Stochastic Processes and their Applications (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2202.01250

zbMATH Keywords

supermartingaleslaw of the iterated logarithmheavy-tailed distributionssequential analysisconfidence sequences

Mathematics Subject Classification ID

Martingales with discrete parameter (60G42) Sequential estimation (62L12)

Related Items (1)

Game-theoretic statistics and safe anytime-valid inference

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Sub-Gaussian mean estimators
• \(L_p\)-version of the Dubins-Savage inequality and some exponential inequalities
• Exponential inequalities for sums of weakly dependent variables
• On confidence sequences
• A note on confidence sequences in multiparameter exponential families
• Challenging the empirical mean and empirical variance: a deviation study
• Time-uniform, nonparametric, nonasymptotic confidence sequences
• A generalized Catoni's M-estimator under finite \(\alpha\)-th moment assumption with \(\alpha \in (1,2)\)
• Time-uniform Chernoff bounds via nonnegative supermartingales
• Mean estimation and regression under heavy-tailed distributions: A survey
• The Nonexistence of Certain Statistical Procedures in Nonparametric Problems
• Law of the Iterated Logarithm and Invariance Principle for M-Estimators
• Always Valid Inference: Continuous Monitoring of A/B Tests
• A TCHEBYCHEFF-LIKE INEQUALITY FOR STOCHASTIC PROCESSES
• Bandits With Heavy Tail
• CONFIDENCE SEQUENCES FOR MEAN, VARIANCE, AND MEDIAN
• On the law of the iterated logarithm

This page was built for publication: Catoni-style confidence sequences for heavy-tailed mean estimation