Dynamic height growth models for highly productive pedunculate oak (Quercus robur L.) stands: explicit mapping of site index classification in Serbia

Table 5 Indicators of the model’s goodness-of-fit

Model	Fitting statistics					Validation statistics			Growth elements
Model	\(\overline{e }\)	\({\text{RMSE}}\)	\({{R}_{{\text{adj}}}}^{2}\)	d-w	\(AIC\)	Bias	\({\text{MAD}}\)	\({\text{MEF}}\)	Asy	CAI	Age
M1	0.0631	1.1195	0.9676	1.98	3793.96	− 0.0099	0.2439	0.9975	40.4	1.13	11
M2	0.0388	1.0973	0.9707	1.98	3898.04	0.0076	0.2457	0.9975	40.1	1.69	1
M3	0.0387	1.0972	0.9707	1.98	3897.97	0.0064	0.2460	0.9975	40.0	1.79	1
M4	0.0592	1.2588	0.9614	2.01	3970.89	0.0100	0.2440	0.9975	42.1	1.53	7
M5	0.0481	1.1879	0.9656	1.99	3942.91	0.0008	0.2474	0.9975	41.3	1.37	5

\(\overline{{\varvec{e}} }\) estimation bias, \(RMSE\) root mean square error, \({{R}_{adj}}^{2}\) adjusted coefficient of determination, \(AIC\) Akaike information criterion, (d-w) Durbin-Watson statistics; validation statistics: \(MAD\) mean absolute difference, \(MEF\) model estimation efficiency) and growth elements for assessment of biological realism of the models (Asy asymptotes, CAI current annual height increment, Age time of culmination

ISSN: 1297-966X