Chapter 5 and 6 Check Your Understanding
Chapter 5
1. The forecasting staff for the Prizer Corporation has developed a model to predict sales of its air-cushioned ride snowmobiles. The model specifies that sales S vary jointly with disposable income Y and the population between age 15 and 40, and inversely with the price of the snowmobiles P. Based on past data, the best estimate of this relationship is s=k yz/p
Where K has been estimated (with past data) to equal 100
A. If Y = $11,000, Z = $1,200 and P =$20,000 what value would you predict for S?
S = k(YZ)/P =100(11,000*1200)/20,000 = 66,000
B. What happens if P is reduced to $17,500?
S = 100(11,000 /1,200)/17,500=$75,429
C. How would you go about developing a value for K?
You would need the time series data on S, Y, Z and P. You would use the regression techniques to get an estimate K. The double log entry system would be used to get the intercept.
D. What are the potential weaknesses of this model?
Sales in the period t+1 are unknown before t+1 so estimates would have to be made for St, Yt, Zt and Pt.
5. A firm experienced the demand shown in the following table.
A. Fill in the table by preparing forecasts based on a five year moving average, a three year moving average and exponential smoothing (with a w = 0.9 and a w=0.3). Year Demand 5 year 3 year Ex Smoothing Ex Smoothing 0.9 0.3
2000 800 xxx xxx xx xx
2001 925 xxx xxx 800 800
2002 900 xxx xxx 912.50 837.50
2003 1025 xxx 875 901.25 856.25
2004 1150 xxx 950 1012.63 906.88
2005 1160 960 1025 1136.26 979.81
2006 1200 1032 1112 1157.63 1033.87
2007 1150 1087 1170 1195.76 1083.71
2008 1270 1137 1170 1154.58 1103.60
2009 1290 1186 1207 1258.46 1153.52
2010 1214 1237 1286.85 1194.46

Using the forecasts from 2005 through 2009 compare the accuracy of each of the forecasting methods based on RMSE criterion. = 134.1 would be the