Predicting crown width and length using nonlinear mixed-effects models: a test of competition measures using Chinese fir (Cunninghamia lanceolata (Lamb.) Hook.)

Table 10 Comparisons among optimum basic model, optimum basic model with optimum competition index and optimum mixed-effects model for crown width and length, respectively

Equation	Adjust R²	− 2LL (df)	AIC	BIC	Test	L.Ratio	p value	NMSE	PRESS
(10)	0.4620	2552.41 (3)	2558	2574				0.5505	48.4941
(12)	0.5517	2211.93 (4)	2220	2241	(12) vs. (10)	340.48	< 0.0001	0.4390	38.7194
(16)	0.7469	1548.54 (9)	1567	1614	(16) vs. (12)	663.39	< 0.0001	0.2651	23.3120
(11)	0.4900	5346.14 (3)	53,527	5368				0.5302	332.3337
(13)	0.5724	5119.63 (4)	5128	5149	(12) vs. (14)	226.51	< 0.0001	0.4559	285.3949
(17)	0.6301	4780.25 (9)	4798	4846	(14) vs. (18)	339.38	< 0.0001	0.3861	221.7118

Adjust R², adjusted coefficient of determination; − 2LL (df), − 2 log-likelihood (degree of freedom); AIC, Akaike information criterion; BIC, Bayesian information criterion; Test, likelihood ratio test; L.Ratio, the value of likelihood ratio test; NMSE, normalized mean square error; PRESS, predicted the error sum of squares

ISSN: 1297-966X