Table 3.
Model accuracy statistic results for the exponential smoothing methods
| Aluminum | Copper | Lead | Iron | Nickel | Tin | Zinc | ||
|---|---|---|---|---|---|---|---|---|
| Holt’s model | RMSE | 328.619 | 1120.858 | 330.834 | 21.587 | 5035.768 | 3688.866 | 512.261 |
| MAE | 251.683 | 780.696 | 228.594 | 12.769 | 3235.982 | 2477.906 | 304.111 | |
| MAPE | 14.117 | 17.833 | 19.785 | 25.301 | 23.787 | 20.896 | 16.862 | |
| Brown’s model | RMSE | 352.442 | 1232.884 | 349.641 | 21.859 | 5460.880 | 3854.318 | 559.443 |
| MAE | 264.000 | 884.242 | 238.780 | 12.385 | 3594.080 | 2385.927 | 346.252 | |
| MAPE | 14.649 | 18.476 | 18.962 | 16.794 | 26.006 | 15.902 | 18.366 | |
| Damped trend model | RMSE | 334.208 | 1140.284 | 335.379 | 21.072 | 5124.104 | 3753.167 | 520.936 |
| MAE | 251.547 | 779.263 | 221.755 | 11.702 | 3240.150 | 2450.981 | 306.875 | |
| MAPE | 14.106 | 17.759 | 18.674 | 15.686 | 23.795 | 20.296 | 16.746 | |
The significance of the “bold” is indicated as the smallest value in the text