Question

Multiple Choice

1. The high level of sales that department stores experience during Christmas is an example of what component of a time series?
   1. Trend       
   2. Cyclical    
   3. Seasonal
   4. Irregular

Questions 2 – 5 refer to the following time series:

Year: 0 1 2 3 4   

Sales: 12 14 19 23 18

2. The value of the three-year centered moving average that corresponds to year 1 will be

1. 15.            
2. 14.            
3. 18.7.         
4. 13.3.

3. The value of the four-year centered moving aver­age that corresponds to year 2 will be

1. 16.50.        
2. 21.00.
3. 17.75.
4. 18.25.

4. For the smoothing constant α = 0.6, the expo­nentially smoothed value for year 2 will be

1. 14.3.                    
2. 15.7.                    
3. 16.4.                    
4. 16.7.

5. For the smoothing constant α = 0.6, the expo­nentially smoothed forecast for year 2 will be

1. 12.0.                    
2. 13.2.                    
3. 13.6.                    
4. 14.0.

6. In using exponential smoothing to dampen the fluctuations in a time series, for which of the fol­lowing values of the smoothing constant will the smoothed series "catch up" most quickly when­ever the original time series changes directions?

1. α = 0.1    
2. α = 0.4
3. α = 0.7
4. α = 0.9

7. In determining monthly seasonal indexes, the first step is to construct a centered moving aver­age with a period of N = ____ months.

1. 2           
2. 4           
3. 12
4. 24

8. In determining monthly seasonal indexes for sales, the sum of the 12 mean values for sales as a percentage of the moving average is 1187. To get the seasonal indexes each monthly meat must be multiplied by

1. 1200/1187.         
2. 1187/1200.
3. 100.
4. (1200 + 1187)/2.

9. For the trend equation ŷ = 10(2) ^x, with x = 0 for 1988, the forecast for 1994 will be

1. 30.             
2. 640.           
3. 120.           
4. 810.

10. Compared to the MAD criterion, the MSE crite­rion

1. is less applicable to time series that have a negative trend.
2. more heavily penalizes larger estimation errors.
3. is rarely appropriate for evaluating the fit of an estimation equation.
4. There are no differences between these two criteria.