An EWMA-based Bounded Adjustment Scheme With Adaptive Noise Variance Estimation

摘要

Bounded-adjustment schemes have been developed to adjust processes subject to stochastic disturbances for the purpose of minimizing a criterion that includes adjustment costs. These schemes were generally formulated under the assumption that the standard deviation of the inherent white-noise component of the disturbances was known. This paper presents a method for applying a bounded-adjustment scheme to continuous or short-run processes in which the standard deviation of the white noise is unknown or is subject to change. We show through Monte Carlo simulation that this control method may be applied to both nonstationary and stationary processes.