Description of Forecasting Problem 1 The U.S. Census Bureau publishes data on factory orders for all manufacturing, durable goods, and nondurable goods…

Description of Forecasting

Problem 1
The U.S. Census Bureau publishes data on factory orders for all manufacturing, durable goods, and nondurable goods industries. Shown here are factory orders in the United States from 1987 through 1999 ($ billion).

a) Use these data to develop forecasts for the years 1992 through 1999 using a 5-year moving average.

b) Use these data to develop forecasts for the years 1992 through 1999 using a 5-year weighted moving average. Weight the most recent year by 6, the previous year by 4, the year before that by 2, and the other years by 1.

c) Which method result is more suitable for forecasting factory orders? Hint: Compare the two methods based on Mean Absolute Deviation (MAD)?

Year Factory Orders ($ billion)
1987 2,512.7
1988 2,739.2
1989 2,874.9
1990 2,934.1
1991 2,865.7
1992 2,978.5
1993 3,092.4
1994 3,356.8
1995 3,607.6
1996 3,749.3
1997 3,952.0
1998 3,949.0
1999 4,137.0

Problem 2
The following data list worldwide shipments of personal computers (in thousands) according to Dataquest.

Year Shipments (in thousands)
1990 23,738
1991 26,966
1992 32,411
1993 38,851
1994 47,894
1995 60,171
1996 71,065
1997 82,400
1998 97,321

a) Use exponential smoothing to determine the forecast of shipments for the year 1999. Use the actual shipments for 1990 as the starting forecast for 1991. Use a smoothing constant of a = 0.4.

b) Plot the data, fit a trend line, and discuss the strength of prediction of the regression model.

c) Use the regression model to predict the shipments for the year 1999.

d) Compare the two forecasts. Which forecast would you prefer to use and why?