Table 4. Accuracy measures of the models based on a split in training set (January 2015 to February 2019) and test sets (March 2019 to December 2019) for the Spanish data.
| ME | RMSE | MAE | MPE | MAPE | ACF1 | Theil’s U | |
|---|---|---|---|---|---|---|---|
| Training set (ARIMA) | 5.92 | 1907.92 | 1153.94 | -0.08% | 3.08% | -0.02 | 0.65 |
| Test set (ARIMA) | -261.03 | 1271.11 | 797.91 | -0.74% | 2.33% | -0.18 | |
| Training set (ETS) | 20.76 | 1737.65 | 1178.08 | -0.11% | 3.15% | 0.45 | 0.49 |
| Test set (ETS) | -173.02 | 1010.51 | 684.69 | -0.44% | 1.97% | -0.20 | |
| Training set (ALNB) | -81.88 | 3933.45 | 2788.56 | -1.06% | 7.68% | 0.08 | 1.34 |
| Test set (ALNB) | -1210.22 | 2803.56 | 2349.11 | -3.73% | 6.91% | 0.31 |