The sufficient and necessary conditions of the strong law of large numbers under the sub-linear expectations

TL;DR

This paper establishes the necessary and sufficient conditions for the strong law of large numbers under sub-linear expectations, extending classical results to non-additive probability frameworks without assuming continuity of capacities.

Contribution

It provides a comprehensive characterization of the strong law of large numbers under sub-linear expectations, including conditions for convergence of series and a probabilistic proof of the weak law.

Findings

Necessary and sufficient conditions for SLLN under sub-linear expectations.

Conditions for convergence of infinite series of independent variables.

Extension of results to original random variables without additional assumptions.

Abstract

In this paper, by establishing a Borel-Cantelli lemma for a capacity which is not necessarily continuous, and a link between a sequence of independent random variables under the sub-linear expectation and a sequence of independent random variables on $\mathbb R^{\infty}$ under a probability, we give the sufficient and necessary conditions of the strong law of large numbers for independent and identically distributed random variables under the sub-liner expectation, and the sufficient and necessary conditions for the convergence of an infinite series of independent random variables, without any assumption on the continuity of the capacities. A purely probabilistic proof of a weak law of large numbers is also given. In the version 1, there are errors in the proof of Lemma 2.1 and 2.2. Version 2 corrected the errors under additional conditions, but Corollaries 3.1-3.4 are only shown for the copy of the random variables in a new sub-linear expectation space. In this version, we show that Corollaries 3.1-3.4 remain true for the original random variables and the results for the copy are just special cases.