Table 5 Summary of the number predictors and parameters, calibration constants and goodness-of-fit statistics of the models fitted. p is the number of predictors included in the model an R is the relative overfitting rate
Model | \(R^2\) | p | Predictors | NRMSE\(_{632+}\) | \(R^2_{632+}\) | R |
|---|---|---|---|---|---|---|
GLM | 0.414 | 3 | BIO7, BIO8, BIO9 | 0.259 | 0.136 | 0.804 |
Random Forest | 0.802 | 3 | BIO1, BIO2, BIO3 | 0.216 | 0.265 | 0.724 |
Boosted Trees | 0.657 | 3 | BIO1, BIO2, BIO3 | 0.212 | 0.268 | 0.671 |
Cubist | 0.539 | 3 | BIO1, BIO2, BIO3 | 0.206 | 0.285 | 0.535 |
ABC | 0.339 | 6 | BIO4, BIO5, BIO6, BIO7, BIO8, BIO9 | 0.235 | 0.203 | 0.512 |
IP1 | 0.372 | 9 | BIO3, BIO4, BIO9, BIO2, BIO7, BIO12, BIO14, BIO18, BIO19 | 0.205 | 0.292 | 0.332 |
IP2 | 0.272 | 9 | BIO7, BIO8, BIO9, BIO2, BIO4, BIO12, BIO14, BIO18, BIO19 | 0.215 | 0.192 | 0.311 |
MMG1 | 0.384 | 5 | BIO3, BIO4, BIO9, BIO2, BIO7 | 0.204 | 0.298 | 0.313 |
MMG2 | 0.351 | 5 | BIO7, BIO8, BIO9, BIO2, BIO4 | 0.223 | 0.187 | 0.581 |
MMG3 | 0.375 | 5 | BIO7, BIO8, BIO9, BIO3, BIO4 | 0.218 | 0.219 | 0.529 |