On prediction of individual sequences

DOI10.1214/aos/1017939242zbMath0961.62081MaRDI QIDQ1583891

Nicolò Cesa-Bianchi, Gábor Lugosi

Publication date: 5 June 2001

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1214/aos/1017939242

zbMATH Keywords

empirical processes; finite-state machines; covering numbers; universal prediction; prediction with experts; absolute loss

Mathematics Subject Classification ID

62M20: Inference from stochastic processes and prediction

60G25: Prediction theory (aspects of stochastic processes)

Related Items

Unnamed Item, Internal regret in on-line portfolio selection, Internal regret in on-line portfolio selection, Online forecast combinations of distributions: worst case bounds, Sensor networks: from dependence analysis via matroid bases to online synthesis, Regularization in statistics, Predicting a binary sequence almost as well as the optimal biased coin, Prediction and dimension, Model selection by bootstrap penalization for classification, Fast learning rates in statistical inference through aggregation

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• A game of prediction with expert advice
• The weighted majority algorithm
• Sphere packing numbers for subsets of the Boolean \(n\)-cube with bounded Vapnik-Chervonenkis dimension
• Asymptotics via empirical processes. With comments and a rejoinder by the author
• Empirical processes and applications: An overview. (With discussion)
• Weighted sums of certain dependent random variables
• Majorizing measures: The generic chaining
• On the best constants in the Khinchin inequality
• Universal prediction of individual sequences
• Universal prediction
• Probability Inequalities for Sums of Bounded Random Variables
• On the Uniform Convergence of Relative Frequencies of Events to Their Probabilities