Table 3.
Performances of the eight regression models for accrual percentage prediction
| Regression model | Mean validation RMSE (SD) | Mean train RMSE (SD) |
|---|---|---|
| Ridge | 28.19 (2.75) | 23.89 (0.29) |
| Lasso | 27.83 (2.43) | 26.08 (0.29) |
| Decision Tree | 25.67 (3.17) | 20.96 (0.94) |
| Random Forest | 21.54 (2.67) | 16.21 (0.44) |
| AdaBoost | 21.11 (2.57) | 14.98 (0.39) |
| XGBoost | 20.64 (2.69) | 15.13 (0.28) |
| LightGBM | 20.53 (2.72) | 15.03 (0.31) |
| CatBoost | 20.31 (2.53) | 14.56 (0.27) |
AdaBoost = adaptive boosting; CatBoost = categorical boosting; Lasso = least absolute shrinkage and selection operator; LightGBM = light gradient boosted machine; RMSE = root mean square error; SD = standard deviation; XGBoost = eXtreme gradient boosting.