Table 3 Results of direct implementation of predictive Models.
Final Models | \(\:{\varvec{R}}^{2}\) Score | MSE | RMSE | MAE | EVS | MAPE | SMAPE | Max Error |
|---|---|---|---|---|---|---|---|---|
Decision Tree | 0.99699 | 1926.219 | 43.88872 | 27.33363 | 0.99701 | 0.00733 | 0.73071 | 178 |
Random Forest + LR | 0.99729 | 1734.260 | 41.64445 | 23.72606 | 0.9973 | 0.00661 | 0.65524 | 311.23768 |
kNN + LR | 0.99252 | 4792.8369 | 69.23032 | 30.4877 | 0.99252 | 0.00924 | 0.89398 | 691.48382 |
LightGBM | 0.95415 | 29386.438 | 171.42473 | 79.29507 | 0.95420 | 0.02117 | 2.05022 | 1008.3917 |
Gradient Boosting + LR | 0.99471 | 3389.8957 | 58.22281 | 42.34641 | 0.99471 | 0.01175 | 1.17329 | 340.01416 |
AdaBoost + LR | 0.94458 | 35517.396 | 188.46059 | 148.06786 | 0.94485 | 0.03955 | 3.93416 | 626.4249 |
CatBoost | 0.99876 | 791.07331 | 28.12602 | 16.99067 | 0.99876 | 0.00458 | 0.45613 | 182.87230 |
XGBoost | 0.99893 | 681.92540 | 26.11370 | 16.95128 | 0.99893 | 0.00455 | 0.45442 | 121.06347 |
MLP + LR | 0.99786 | 1370.7032 | 37.02300 | 21.20976 | 0.99786 | 0.00569 | 0.57347 | 331.81693 |
GRU | 0.99826 | 1113.1856 | 33.36444 | 22.83463 | 0.99842 | 0.00596 | 0.59937 | 174.52637 |
LSTM + LR | 0.99857 | 917.79771 | 30.29518 | 22.05073 | 0.99857 | 0.00596 | 0.59515 | 104.54755 |
Autoencoders + LR | 0.99921 | 509.21466 | 22.56579 | 13.29329 | 0.99921 | 0.00356 | 0.35602 | 114.69256 |
