Table 3. Relationship between real and predicted range sizes based on a linear regression model.
a is the intercept, b is the slope of the regression. R2 is the coefficient of determination. R2 values higher than 0.8 are highlighted. Higher R2 associated with a≈0 and b≈1 denotes best models for range-size prediction.
| LPT | Balance | ||||||
|---|---|---|---|---|---|---|---|
| Environm. Variables |
Modelling procedures |
a | b | R2 | A | b | R2 |
| All | ES | 107.4087 | 0.7586 | 0.974 | 107.4087 | 0.7586 | 0.974 |
| All | Mahal | -103.1712 | 0.9338 | 0.954 | 129.8441 | 0.8254 | 0.943 |
| All | MXLQ | 261.7759 | 1.5257 | 0.955 | 584.5047 | 1.1697 | 0.958 |
| All | MXDEF | 169.613 | 1.3059 | 0.961 | 590.3274 | 1.0117 | 0.962 |
| All | SVM | 145.9529 | 1.5522 | 0.939 | 716.1815 | 1.1428 | 0.949 |
| 6 PCA | ES | 234.4172 | 1.0185 | 0.972 | 234.4172 | 1.0185 | 0.972 |
| 6 PCA | Mahal | 956.7514 | 1.4758 | 0.943 | 1071.594 | 1.1754 | 0.942 |
| 6 PCA | MXLQ | 484.8208 | 1.4689 | 0.940 | 773.1833 | 1.16 | 0.959 |
| 6 PCA | MXDEF | 416.2603 | 1.4710 | 0.945 | 752.6322 | 1.1566 | 0.958 |
| 6 PCA | SVM | 676.0404 | 1.7828 | 0.901 | 1261.578 | 1.1252 | 0.919 |
| 4 PCA | ES | 291.2412 | 1.1855 | 0.971 | 291.2412 | 1.1855 | 0.971 |
| 4 PCA | Mahal | 628.8269 | 1.7099 | 0.926 | 1142.131 | 1.2586 | 0.954 |
| 4 PCA | MXLQ | 682.4333 | 1.5067 | 0.943 | 927.4211 | 1.2327 | 0.954 |
| 4 PCA | MXDEF | 658.1231 | 1.5100 | 0.940 | 922.4394 | 1.237 | 0.953 |
| 4 PCA | SVM | 1040.6258 | 1.7033 | 0.894 | 1306.712 | 1.2166 | 0.936 |