Ww2-secure.justanswer.com



I wanted to include some notes to explain the Excel charts I’m sending you. I used the Holt-Winters double exponential smoothing technique listed here: and here 1 (See Tabs Problem 1 Chart, Problem 1 Data)I put in initial values for α=0.1 and β=0.9 per your class notes and then used the Excel Solver feature to get a minimum value for the Mean Square Error (MSE) of the smoothed value compared to the Demand value. The optimal values are α=0.8747 and β=0.0065 which gave a MSE of 469.34 (originally 1055).At this point the smooth curve becomes the basis for a straight-line forecast of the familiar slope intercept form (y=mx+b). In this particular case beginning for the 2005 forecast it is: Forecast=533.37+period*14.03 where period is the number of periods of the forecast (period (2005)=1, period (2006)=2)Problem 2 (See Tabs Problem 2 Chart, Problem 2 Data)I put in initial values for α=0.1 and β=0.9 per your class notes and then used the Excel Solver feature to get a minimum value for the Mean Square Error (MSE) of the smoothed value compared to the Demand value. The optimal values are α=0.0000 and β=0.8121 which gave a MSE of 43.1061 (originally 47.1727).At this point the smooth curve becomes the basis for a straight-line forecast of the familiar slope intercept form (y=mx+b). In this particular case beginning for the 2012 Q1 forecast it is: Forecast=43.000+period*1.818 where period is the number of periods of the forecast (period (2012 Q1)=1, period (2012 Q2)=2, etc.)Don’t forget to put your name in the file name ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download