Partial preview of the text
Download SCMG-305-Global-Demand-Management-Assignment and more Assignments Business Systems in PDF only on Docsity!
STANDARD STATISTICAL ERROR TERMS Forecast performance methods are also used to measure and com- pare the accuracy of various forecasting methods. The measures we use to determine the effectiveness of our forecasts are interchangeable in helping to determine the appropriate method(s) to deploy. Before describing the methods, I will define some terms and data require- ments required to measure forecast performance. As we discussed in Chapter 3, our purpose as practitioners is to minimize error in the forecast. Those errors are defined as the differ- ence between what was forecasted (expected sales) for a specific time period and the actual (sales) value. The basic measure of a forecast model can be written as: Error =e, =A, fy where e, — error for time period t A, — actual sales tor time period t F, — forecast for time period t An error value is always associated with each observation for which there is both an actual and a predicted value. To simplify the manipulation of expressions involving the adding of many numbers, 
