Sales of iPhone 6 have been going well for a local electroni
Sales of iPhone 6 have been going well for a local electronic store. These sales for the past fourteen weeks are shown below:
a) Use exponential smoothing constant of 0.2 to forecast the sales of iPhone6i. Assuming the forecast for week 1 is 35 iphone 6s. What is the MAD of the model?
b) Forecast the sales for week 15 using a 5-week weighted moving average. Use 5,4,3,2, and 1 for the weights of the most recent, second most recent, third most recent, fourth most recent and fifth most recent periods, respectively. What is the MAD of the model? What is the forecast value for week 15?
c) Would you prefer to use the exponential smoothing model or the 5-week weighted model? Explain your answer.
| Week | Sales |
| 1 | 35 |
| 2 | 36 |
| 3 | 30 |
| 4 | 23 |
| 5 | 38 |
| 6 | 15 |
| 7 | 20 |
| 8 | 27 |
| 9 | 18 |
| 10 | 24 |
| 11 | 30 |
| 12 | 32 |
| 13 | 34 |
| 14 | 36 |
Solution
The mean absolute deviation for exponential smoothing is giving a smaller absolute deviation value, which means it is a better forecast value. Thus, exponential smoothing might be preferred over weighted average method in this case.
Hope this helps. ask if you have any doubts
| alpha = | 0.2 | ||||
| Week | Sales | Forecast | MAD | ||
| Exponential | Wt. Average | Exponential | Wt. Average | ||
| 1 | 35 | 35 | 0.00 | ||
| 2 | 36 | 35.000 | 1.00 | ||
| 3 | 30 | 35.200 | 5.20 | ||
| 4 | 23 | 34.160 | 11.16 | ||
| 5 | 38 | 31.928 | 6.07 | ||
| 6 | 15 | 33.142 | 31.933 | 18.14 | 16.933 |
| 7 | 20 | 29.514 | 26.133 | 9.51 | 6.133 |
| 8 | 27 | 27.611 | 23.333 | 0.61 | 3.667 |
| 9 | 18 | 27.489 | 23.933 | 9.49 | 5.933 |
| 10 | 24 | 25.591 | 21.733 | 1.59 | 2.267 |
| 11 | 30 | 25.273 | 21.867 | 4.73 | 8.133 |
| 12 | 32 | 26.218 | 24.933 | 5.78 | 7.067 |
| 13 | 34 | 27.375 | 27.667 | 6.63 | 6.333 |
| 14 | 36 | 28.700 | 30.267 | 7.30 | 5.733 |
| 15 | 30.160 | 33.067 | |||
| MAD | 6.23 | 6.91 |

