Free Access

Table II

Parameter estimates, their standard errors and goodness of fit statistics for the models developed to predict total above ground biomass. Values of root mean square error (RMSE) and the Akaike information criterion (AIC) can only be used for comparisons with the same class of model (i.e. logarithmic transformed models or weighted arithmetic models).

Equation No. Parameter estimates RMSE AIC FI R2

β 0 β 1 β 2 β 3 β 4
Logarithmic transformed models
3 –1.9400 2.1824 0.2917 323.5 13.72 0.92
(0.3726) (0.1236)
4 –0.5816 1.1240 0.2030 0.2332 44.8 10.97 0.95
(0.3667) (0.1212) (0.0108)
5 –2.1004 1.8594 0.4370 0.2709 234.0 12.74 0.93
(0.3672) (0.1282) (0.1310)
6 –0.9753 1.0241 0.1811 0.3399 0.2268 13.6 10.67 0.95
(0.3375) (0.1127) (0.0110) (0.1100)
7 –0.9069 1.2273 0.1411 –0.0078 0.0840 0.2266 11.4 10.66 0.95
0.3452 0.1357 0.0221 0.2126 0.0400
8 –2.4150 0.8186 0.2768 246.9 13.02 0.93
0.4170 0.0455
9 –1.1602 0.4859 0.0219 0.2383 58.2 11.20 0.95
Weighted arithmetic models
13 0.1291 2.2718 0.3052 5436 16.39 0.88
0.0080 0.0435
14 0.1039 0.8229 0.2409 5406 16.12 0.94
(0.0113) (0.0071)
15 0.0944 1.5286 0.2433 5458 16.76 0.95
(0.0111) (0.0214)
16 0.0994 0.4206 0.2694 5347 15.38 0.95
(0.0101) (0.0213)
17 0.0977 2.0362 0.3845 0.2734 5349 15.38 0.95
(0.0100) (0.0472) (0.0555)
18 69.4057 0.2877 0.1007 5721 19.59 0.78
(45.5534) (0.0053)
19 2.0597 0.0185 0.2190 5571 18.18 0.94
(0.9392) (0.0010)
20 0.7013 0.0807 0.1620 5237 13.97 0.97
(0.3527) (0.0057)