3) Develop forecasts for June through October using these techniques: moving average of two period, simple exponential smoothing with an alpha of 0.8, and Holt's method. For the exponential smoothing model assume that the forecast for May is the actual demand for May Comment on the use of these three methods to generate a forecast in this situation. Answer: The moving average of two periods results in these forecasts: Actual Month June July August September October 8 29 30 MA2) 41.5 48.5 60 74,5 82 89 Fg = 8189 The exponential smoothing with a=0.8 Month Actual Expon.8) June 53 44 July 67 51.2 August 82 63.8 September 96 78.4 October 92.5 For Holt's model the regression equation has an intercept of -2.5 and a trend component of 10.3. Beta was used as 0.1 and alpha 0.2. Month Actual Forecast Error Level Trend June 53 59.1 6,1 57.8 10.1 July 67 58.0 1.0 67.8 10.1 August 82 77.9 -4.1 78.7 10.2 September 88.9 -7.1 90.3 10.3 October 100.7 96 The data show a strong trend, so use of the simple moving average or exponential smoothing will just result in forecasts that lag behind the actual demand. Holt's model can capture the trend and provide a much more accurate forecast Diff 2 Topic: 7.5 Time-Series Forecasting Methods AACSB: Analytical thinking Objective: LO 7.3: Forecast demand in a supply chain given historical demand data using time- series methodologies, a 23 Cost Accounting: A Managerial Emphasis