New Zealand Journal of Forestry Science

Table 4 Performance metrics of linear models estimating MOE of standing trees. Italicised values indicate the recommended model; bold values indicate the best value(s) in each column

From: Validation of models predicting modulus of elasticity in Douglas-fir trees, boles, and logs

	Calibration dataset						Validation dataset			Comments
Fixed effects^a	R ² _marg	R ² _cond	AIC	RMSE	MAPE	TP rate	RMSE	MAPE	TP rate
				(GPa)	(%)	(%)	(GPa)	(%)	(%)
V ²	0.36	0.45	465	1.3	10	58	1.5	11	46
D	0.18	0.49	493	1.6	12	31	1.2	7	55
ρ	0.18	0.44	487	1.5	11	36	1.2	8	27
T	0.32	0.62	483	1.5	13	53	1.2	9	82	Best 1-variable
E	0.43	0.54	443	1.2	9	67	1.4	10	73
V ² + ρ	0.44	0.56	438	1.2	9	67	1.2	9	64	Similar to E
V ² + T	0.45	0.58	449	1.2	10	60	1.3	10	82
V ² + D	0.37	0.50	457	1.3	10	56	1.3	10	64
D + H	0.33	0.64	480	1.5	12	56	1.3	9	82	Similar to T
D + ρ	0.31	0.61	449	1.4	10	42	1.2	9	55
H + ρ	0.21	0.59	479	1.6	13	29	1.4	10	36
E + D	0.45	0.57	434	1.2	9	64	1.3	9	64
E + T	0.51	0.60	429	1.1	9	69	1.3	10	91
E + D + H	0.51	0.61	428	1.1	9	69	1.3	10	91
V ² + D + H	0.46	0.60	447	1.2	10	60	1.4	11	82
V ² + ρ + D	0.46	0.61	422	1.1	8	64	1.1	8	73	Best AIC
V ² + ρ + T	*0.51*	0.61	423	*1.1*	8	71	*1.1*	8	91	Best validation

^aAll variables are as defined in Table 2

Back to article page