An approach to complete convergence theorems for dependent random fields via application of Fuk Nagaev inequality

Abstract

Let $\{ X_{\bf n}, {\bf n}\in \mathbb{N}^d \}$ be a random field i.e. a family of random variables indexed by $\mathbb{N}^d $, $d\ge 2$. Complete convergence, convergence rates for non identically distributed, negatively dependent and martingale random fields are studied by application of Fuk-Nagaev inequality. The results are proved in asymmetric convergence case i.e. for the norming sequence equal $n_1^{\alpha_1}\cdot n_2^{\alpha_2}\cdot\ldots\cdot n_d^{\alpha_d}$, where $(n_1,n_2,\ldots, n_d)=\mathbf{n} \in \mathbb{N}^d$ and $\min\limits_{1\leq i \leq d}\alpha_i \geq \frac{1}{2}.$