Complete convergence for arrays of rowwise widely orthant dependent random variables and its applications

Abstract

In this paper, we study complete convergence for arrays of rowwise widely orthant dependent (WOD) random variables. Some general results on complete convergence for arrays of rowwise WOD random variables are established, which generalize the corresponding ones for extended negatively dependent random variables. As applications, we present some corollaries on complete convergence for WOD random variables, and present a result on complete consistency for the estimator in a nonparametric regression model based on WOD errors.