TABLE 4.

Indica 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	100	55.56	38	63.33	1.027	0.00	0.511	0.00	3,023.024	0.00	0.364	0.00
M1	83	46.11	24	40.00	1.041	−1.40	0.513	0.22	2,993.455	0.99	0.375	−2.99
M2	75	41.67	26	43.33	1.040	−1.25	0.510	−0.25	3,012.109	0.36	0.432	−15.89
M3	75	41.67	27	45.00	1.082	−5.16	0.514	0.41	2,754.294	9.76	0.333	9.29
M4	76	42.22	27	45.00	1.084	−5.26	0.512	0.19	2,731.500	10.67	0.372	−2.25
M5	81	45.00	20	33.33	1.121	−8.46	0.512	0.13	2,661.258	13.59	0.327	11.19
M6	83	46.11	29	48.33	1.041	−1.37	0.514	0.44	2,548.352	18.63	0.293	24.07
M7	86	47.78	29	48.33	1.043	−1.56	0.512	0.18	2,634.676	14.74	0.339	7.30
M8	108	60.00	37	61.67	1.029	−0.27	0.513	0.22	2,794.690	8.17	0.299	21.45
M9	24	13.33	2	3.33	1.173	−12.46	0.072	−85.98	12,217.216	−75.26	2.047	−82.24
M10	32	17.78	9	15.00	1.122	−8.52	0.462	−9.61	7,927.607	−61.87	1.039	−65.01
M11	112	62.22	44	73.33	1.021	0.59	0.512	0.10	3,045.173	−0.73	0.330	10.13
M12	124	68.89	48	80.00	0.976	5.15	0.513	0.27	2,464.826	22.65	0.253	43.84
M13	93	51.67	21	35.00	1.13	−9.38	0.513	0.29	3,252.671	−7.06	0.306	18.62
M14	92	51.11	33	55.00	1.03	−0.44	0.512	0.02	2,780.146	8.74	0.348	4.48