Complete convergence for arrays of rowwise <italic>m</italic><italic>n</italic>-extended negatively dependent random variables and its application.

Abstract.In this article, under some proper and sufficient conditions, we gave complete convergence for weighted sums and maximal weighted sums of arrays of rowwise <italic>m</italic><italic>n</italic>-extended negatively dependent (rowwise <italic>m</italic><italic>n</italic>-END) random variables, which is a new dependent structure. In addition, a relationship between {mn,n≥1} and moment condition for convergence is revealed in a sense. The results obtained in the article generalize some corresponding ones for independent and some dependent random variables. As an application, the strong consistency for the weighted estimator in a non parametric regression model is established. [ABSTRACT FROM AUTHOR]