Remark on the Berry-Esseen Bound for Negatively associated Random Variables
摘要
In this work, we study the accuracy of normal approximation for a specific type of dependent random sequence characterized by a form of negative association and stationarity. Previous studies have achieved certain convergence rates under specific conditions on how dependencies weaken over time and assuming the variables are bounded. Our approach shows that by carefully adjusting how the data is grouped-specifically, modifying the sizes of certain segments used in the analysis-it is possible to achieve a more accurate approximation than some earlier methods.