Determination of the Best Simple Moving Average By Stochastic Processes

Abstract

In this study, we consider one of the most popular technical indicators and try to determine the best fitting simple moving average to a given data. Here we utilize from a general mean reverting stochastic process where the mean is time dependent. We propose an identification algorithm which mainly concentrates on the normality of the residual terms after the data is demeaned from simple moving average and also provide evidence that our algorithm works quite well for determination of the “best” simple moving average.