TABLE 6.

G2F_2014 data. Count of the number of times a model was better than another in terms of Normalized Root Mean Squared Error (NRMSE), both by environments and by traits. Prediction accuracy in terms of NRMSE. b0 denotes the absolute values of the intercept and b denotes the absolute value of 1-slope. Under an ideal model both b0 and b should be equal to zero. Relative efficiency (RE) or each model in percentage was computed regarding model M0 without environmental covariates. When the percentage is positive there is a gain in prediction accuracy regarding M0, while when the percentage is negative there is a loss in terms of prediction accuracy of any model regarding M0.

Model	Env		Trait		NRMSE	NRMSE	Cor	Cor	b0	b0	b	b
	Won	%	Won	%		RE (%)		RE (%)		RE (%)		RE (%)
	Models		Models
M0	454	31.53	25	31.25	4.297	0.00	0.379	0.00	0.053	0.00	0.048	0.00
M1	590	40.97	41	51.25	4.675	−8.08	0.381	0.66	0.746	−92.83	0.014	247.99
M2	466	32.36	15	18.75	5.587	−23.08	0.374	−1.26	1.524	−96.50	0.307	−84.47
M3	515	35.76	32	40.00	7.632	−43.70	0.379	0.00	0.371	−85.61	0.039	23.51
M4	447	31.04	16	20.00	12.259	−64.95	0.332	−12.27	2.588	−97.94	0.392	−87.83
M5	467	32.43	24	30.00	6.569	−34.58	0.377	−0.46	0.392	−86.37	0.045	6.18
M6	576	40.00	45	56.25	5.088	−15.54	0.376	−0.73	0.002	2,126.04	0.042	14.19
M7	438	30.42	13	16.25	9.663	−55.53	0.332	−12.32	2.212	−97.58	0.393	−87.86
M8	424	29.44	14	17.50	7.543	−43.03	0.337	−10.91	2.329	−97.71	0.391	−87.80
M9	422	29.31	21	26.25	4.699	−8.56	0.293	−22.53	0.627	−91.48	0.334	−85.74
M10	504	35.00	34	42.50	5.464	−21.35	0.265	−29.91	8.143	−99.34	0.945	−94.96
M11	509	35.35	23	28.75	8.663	−50.39	0.358	−5.36	2.182	−97.55	0.306	−84.44
M12	553	38.40	31	38.75	7.035	−38.92	0.368	−2.92	1.416	−96.23	0.267	−82.11
M13	608	42.22	41	51.25	6.274	−31.51	0.378	−0.08	0.230	−76.78	0.038	25.46
M14	586	40.69	45	56.25	5.510	−22.02	0.378	−0.11	0.052	1.76	0.042	13.92