In: Statistics and Probability
Consider the following gasoline sales time series. If needed, round your answers to two-decimal digits.
Week | Sales (1,000s of gallons) |
1 | 17 |
2 | 21 |
3 | 19 |
4 | 23 |
5 | 18 |
6 | 16 |
7 | 20 |
8 | 18 |
9 | 22 |
10 | 20 |
11 | 15 |
12 | 22 |
(a) | Show the exponential smoothing forecasts using α = 0.1, and α = 0.2. | |||||||||
|
||||||||||
(b) | Applying the MSE measure of forecast accuracy, would you prefer a smoothing constant of α = 0.1 or α = 0.2 for the gasoline sales time series? | |||||||||
An - Select your answer -α = 0.1 / α = 0.2 smoothing constant provides the more accurate forecast, with an overall MSE of . | ||||||||||
(c) | Are the results the same if you apply MAE as the measure of accuracy? | |||||||||
An - Select your answer -α = 0.1 / α = 0.2 smoothing constant provides the more accurate forecast, with an overall MAE of . | ||||||||||
(d) | What are the results if MAPE is used? | |||||||||
An - Select your answer -α = 0.1α = 0.2 smoothing constant provides the more accurate forecast, with an overall MAPE of . |
Exponential smoothing (α=0.1)
Week | Sales (1,000s of gallons)(Y) | Y^ = ES(0.1) = αAt+(1-α)Ft | ABS Error = |Y-Y^| | Error^2 | Absolute % error = ABS error/Y |
1 | 17 | 17 | 0 | 0 | 0.00% |
2 | 21 | 17 | 4 | 16 | 19.05% |
3 | 19 | 17.4 | 1.6 | 2.56 | 8.42% |
4 | 23 | 17.56 | 5.44 | 29.5936 | 23.65% |
5 | 18 | 18.104 | 0.104 | 0.010816 | 0.58% |
6 | 16 | 18.0936 | 2.0936 | 4.383161 | 13.09% |
7 | 20 | 17.88424 | 2.11576 | 4.47644 | 10.58% |
8 | 18 | 18.095816 | 0.095816 | 0.009181 | 0.53% |
9 | 22 | 18.0862344 | 3.9137656 | 15.31756 | 17.79% |
10 | 20 | 18.47761096 | 1.52238904 | 2.317668 | 7.61% |
11 | 15 | 18.62984986 | 3.629849864 | 13.17581 | 24.20% |
12 | 22 | 18.26686488 | 3.733135122 | 13.9363 | 16.97% |
13 | 18.64017839 | ||||
Average | 2.3540 | 8.4817 | 11.87% |
Exponential smoothing (α=0.2)
Week | Sales (1,000s of gallons)(Y) | Y^ = ES(0.2) = αAt+(1-α)Ft | ABS Error = |Y-Y^| | Error^2 | Absolute % error = ABS error/Y |
1 | 17 | 17 | 0 | 0 | 0.00% |
2 | 21 | 17 | 4 | 16 | 19.05% |
3 | 19 | 17.8 | 1.2 | 1.44 | 6.32% |
4 | 23 | 18.04 | 4.96 | 24.6016 | 21.57% |
5 | 18 | 19.032 | 1.032 | 1.065024 | 5.73% |
6 | 16 | 18.8256 | 2.8256 | 7.9840154 | 17.66% |
7 | 20 | 18.26048 | 1.73952 | 3.0259298 | 8.70% |
8 | 18 | 18.608384 | 0.608384 | 0.3701311 | 3.38% |
9 | 22 | 18.4867072 | 3.5132928 | 12.343226 | 15.97% |
10 | 20 | 19.18936576 | 0.81063424 | 0.6571279 | 4.05% |
11 | 15 | 19.35149261 | 4.351492608 | 18.935488 | 29.01% |
12 | 22 | 18.48119409 | 3.518805914 | 12.381995 | 15.99% |
13 | 19.18495527 | ||||
Average | 2.3800 | 8.2337 | 12.29% |
a)
Exponential | ||
Smoothing | ||
Week | α = 0.1 | α = 0.2 |
13 | 18.6402 | 19.185 |
b)
Exponential | ||
Smoothing | ||
α = 0.1 | α = 0.2 | |
MSE | 8.4817 | 8.2337 |
MSE(α = 0.2) < MSE(α = 0.1)
An α = 0.2 smoothing constant provides the more accurate forecast, with an overall MSE of 8.2337 |
c)
Exponential | ||
Smoothing | ||
α = 0.1 | α = 0.2 | |
MAE | 2.354 | 2.38 |
MAE(α = 0.2) > MAE(α = 0.1)
An α = 0.1 smoothing constant provides the more accurate forecast, with an overall MAE of 2.354
d)
Exponential | ||
Smoothing | ||
α = 0.1 | α = 0.2 | |
MAPE | 11.87% | 12.29% |
An α = 0.1 smoothing constant provides the more accurate forecast, with an overall MAPE of 11.87%
MAPE(α = 0.2) > MAPE(α = 0.1)