Skip to main content

Table 8 MAPE and RMSE results by periods, for overall forecasting accuracy

From: Load forecasting for energy communities: a novel LSTM-XGBoost hybrid model based on smart meter data

Evaluated period MAPE (overall) [%] / RMSE (overall)
SLP LSTM LSTM SM LSTM SM XGB
01.01–30.01. 24.85 / 3.52 19.63 / 3.70 18.25 / 3.27 17.98 / 3.23
31.01.–01.03. 25.64 / 3.87 19.42 / 3.20 25.35 / 3.87 25.04 / 3.83
02.03.–31.03. 24.05 / 3.94 17.82 / 2.79 11.18 / 2.14 11.04 / 2.11
01.04.–30.04. 24.95 / 3.97 25.63 / 3.00 19.40 / 2.83 19.13 / 2.79
01.05.–30.05. 28.03 / 4.71 24.81 / 2.84 23.57 / 2.79 23.61 / 2.79
31.05.–29.06. 27.53 / 4.78 28.50 / 3.33 10.71 / 1.60 10.70 / 1.59
30.06.–29.07. 27.76 / 4.81 16.69 / 1.96 11.00 / 1.41 11.15 / 1.43
30.07.–28.08. 27.01 / 4.87 23.68 / 2.60 20.41 / 2.35 20.39 / 2.35
29.08.–27.09. 28.64 / 4.90 21.61 / 2.68 17.36 / 2.06 17.40 / 2.06
28.09.–27.10. 27.19 / 4.46 16.89 / 2.19 11.29 / 1.73 11.20 / 1.71
28.10.–26.11. 21.50 / 3.21 17.38 / 2.52 15.39 / 2.74 14.99 / 2.67
27.11.–26.12. 26.20 / 3.71 27.63 / 4.13 19.52 / 3.36 19.16 / 3.30
Average 26.11 / 4.23 21.64 / 2.91 16.95 / 2.51 16.81 / 2.49
  1. Best values in respective rows highlighted in bold font and underlined