🤖 AI Summary
This work proposes a novel approach based on online betting strategies to derive empirical Bernstein-type laws of the iterated logarithm (LIL). By constructing betting strategies with guaranteed wealth growth and integrating martingale theory with empirical variance information, the study establishes a rigorous connection between online learning and classical probabilistic limit theorems. This is the first systematic application of an online betting framework to LIL analysis, yielding non-asymptotic bounds that explicitly depend on empirical variance. The resulting bounds not only enhance the applicability of the iterated logarithm law in sequential prediction and statistical inference but also forge a new theoretical bridge between gambling strategies and probabilistic limit theory.
📝 Abstract
This is a verbatim copy of a technical report I wrote in 2017-2018 to obtain the law of the iterated logarithm using the guarantee on the wealth of an online betting strategy.