Table 6 Mean percent of errors, MSE, and mean of errors in classical models, Bayesian averaging model and GBDT algorithm according to the training and the testing datasets
From: Bayesian model averaging for predicting factors associated with length of COVID-19 hospitalization
| Â | Training | Training | ||||
|---|---|---|---|---|---|---|
Fitted models | Mean percentage errors | MSE | Mean of errors | Mean percentage errors | MSE | Mean of errors |
Stepwise | 61.33 | 12.90 | 2.618 | 89.40 | 92.50 | 4.92 |
AIC | 63.40 | 13.30 | 2.691 | 86.29 | 85.83 | 4.58 |
BIC | 61.77 | 13.20 | 2.633 | 87.60 | 85.68 | 4.62 |
Occam’s Window | 54.40 | 10.97 | 2.003 | 60.53 | 75.82 | 3.57 |
MCMC | 55.40 | 10.95 | 2.008 | 61.89 | 78.67 | 3.57 |
GBDT | 40.60 | 5.50 | 1.730 | 87.08 | 83.15 | 4.61 |