On the laws of the iterated logarithm under the sub-linear expectations without the assumption on the continuity of capacities

TL;DR

This paper proves general forms of the law of the iterated logarithm for independent, not necessarily identically distributed variables in sub-linear expectation spaces without requiring capacity continuity, expanding theoretical understanding.

Contribution

It establishes the law of the iterated logarithm under sub-linear expectations without assuming capacity continuity, filling a gap in previous proofs.

Findings

Law of the iterated logarithm holds without capacity continuity assumptions

Provides exponential inequalities for maximum sums of independent variables

Characterizes conditions for i.i.d. variables under sub-linear expectations

Abstract

In this paper, we establish some general forms of the law of the iterated logarithm for independent random variables in a sub-linear expectation space, where the random variables are not necessarily identically distributed. Exponential inequalities for the maximum sum of independent random variables and Kolmogorov's converse exponential inequalities are established as tools for showing the law of the iterated logarithm. As an application, the sufficient and necessary conditions of the law of iterated logarithm for independent and identically distributed random variables under the sub-linear expectation are obtained. In the paper, it is also shown that if the sub-linear expectation space is rich and regular enough, it will have no continuous capacity. The laws of the iterated logarithm are established without the assumption on the continuity of capacities. This revision fills a gap in the proof of Proposition 4.2 of arXiv:2103.01390 under an additional assumption.