Complete q-th moment convergence for the maximum of partial sums of -negatively associated random variables and its application to the EV regression model*

Abstract In this article, we prove the complete q-th moment convergence for the maximum of partial sums of -negatively associated random variables under some general conditions. The results obtained in this article are extensions of previous studies for -negatively associated random variables. In addition, we investigate the strong consistency of the least squares estimator in the simple linear errors-in-variables model based on -negatively associated random variables, and provide some simulations to assess the finite sample performance of the theoretical results.