Table 4.
Model No. | Form of the Model Tested | Fitted Coefficient and Constant | R2 | RMSE | BIC | PRESS | SSE | Bias (PRESS-SSE) |
|
---|---|---|---|---|---|---|---|---|---|
a (SE/p-Value) | b (SE/p-Value) | ||||||||
1 | LA = a + b × L | −83.292 (1.801/***) | 8.510 (0.087/***) | 0.836 | 16.14 | 15,664 | 486,356 | 484,990 | 1366 |
2 | LA = a × eb×L | 14.610 (0.291/***) | 0.086 (0.001/***) | 0.851 | 15.74 | 15,571 | 462,944 | 461,346 | 1598 |
3 | LA = a + b × L2 | 2.227 (0.901/*) | 0.203 (0.002/***) | 0.853 | 15.27 | 15,458 | 435,447 | 434,241 | 1206 |
4 | LA = a + b × W | −61.081 (1.321/***) | 23.575 (0.202/***) | 0.880 | 13.83 | 15,089 | 357,115 | 356,091 | 1024 |
5 | LA = a × eb×W | 19.217 (0.284/***) | 0.230 (0.002/***) | 0.878 | 13.82 | 15,087 | 357,303 | 355,669 | 1634 |
6 | LA = a + b × W2 | 14.025 (0.666/***) | 1.741 (0.014/***) | 0.894 | 12.98 | 14,854 | 314,867 | 313,948 | 919 |
7 | LA = a + b × (L × W) | −0.516 (0.321/ns) | 0.667 (0.002/***) | 0.980 | 5.614 | 11,732 | 58,879 | 58,694 | 185 |
Note: *** p < 0.001; * p < 0.05; ns = not significant; R2 = coefficient of determination; RMSE (cm2) = root mean squared error; BIC = Bayesian information criterion, PRESS = predicted residual error sum of squares; SSE = sum of squared error; Bias = differences between the PRESS and SSE values.