A Generalization of the Petrov Strong Law of Large Numbers

Valery Korchevsky

Published 2014-08-17Version 1

In 1969 V.V.~Petrov found a new sufficient condition for the applicability of the strong law of large numbers to sequences of independent random variables. He proved the following theorem: let $\{X_{n}\}_{n=1}^{\infty}$ be a sequence of independent random variables with finite variances and let $S_{n}=\sum_{k=1}^{n} X_{k}$. If $Var (S_{n})=O (n^{2}/\psi(n))$ for a positive non-decreasing function $\psi(x)$ such that $\sum 1/(n \psi(n)) < \infty$ (Petrov's condition) then the relation $(S_{n}-ES_{n})/n \to 0$ a.s. holds. In 2008 V.V.~Petrov showed that under some additional assumptions Petrov's condition remains sufficient for the applicability of the strong law of large numbers to sequences of random variables without the independence condition. In the present work, we generalize Petrov's results (for both dependent and independent random variables), using an arbitrary norming sequence in place of the classical normalization.