A Note on EWMA Mini paper Taken from Summer 1997 Newsletter
Abstract: Exponentially Weighted Moving Averages (EWMA) has been found to be a powerful and sensitive method, based on exponential smoothing, which can be used as an effective process monitoring technique as well as a useful prediction tool for short term forecasting. The most important advantage of EWMA comes from its flexibility in using data on the history of the process. This is done by selecting a weight factor and effectively using it to achieve the desired balance between the older observations and more recent observations. In part I of this paper the formulae for EWMA are derived, and a simple application involving the forecast for the average weekly production of a particular consumer item is used to explain the concept. In part II a simple and practical method to select the weight factor is presented along with an example involving the forecast of the time-sequenced average daily values and closing values of the NASDAQ stock volumes.
Keywords: Exponentially Weighted Moving Averages (EWMA) - Control charts - Control limits - Cumulative sum control chart (CUSUM)