TABLE 5.

Japonica 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	122	40.67	18	30.00	1.946	0.00	0.584	0.00	102.216	0.00	0.062	0.00
M1	153	51.00	36	60.00	2.029	−4.05	0.585	0.16	183.728	−44.37	0.069	−9.13
M2	92	30.67	7	11.67	3.752	−48.13	0.553	−5.27	151.100	−32.35	0.061	2.09
M3	141	47.00	26	43.33	1.843	5.62	0.584	−0.03	117.624	−13.10	0.067	−6.51
M4	123	41.00	25	41.67	1.858	4.78	0.523	−10.42	90.644	12.77	0.173	−63.86
M5	115	38.33	20	33.33	2.049	−5.00	0.587	0.51	25.585	299.51	0.084	−26.04
M6	131	43.67	21	35.00	2.111	−7.80	0.583	−0.19	46.789	118.46	0.069	−9.66
M7	127	42.33	26	43.33	1.940	0.35	0.526	−9.96	141.541	−27.78	0.147	−57.40
M8	141	47.00	33	55.00	1.786	8.97	0.536	−8.27	344.518	−70.33	0.100	−37.85
M9	107	35.67	20	33.33	1.818	7.04	0.020	−96.49	1793.721	−94.30	1.185	−94.73
M10	138	46.00	34	56.67	1.704	14.25	0.390	−33.19	970.116	−89.46	0.944	−93.39
M11	150	50.00	29	48.33	1.997	−2.51	0.567	−3.02	192.060	−46.78	0.001	4,205.17
M12	178	59.33	41	68.33	1.877	3.70	0.561	−3.91	24.735	313.24	0.061	2.97
M13	185	61.67	42	70.00	1.772	9.84	0.583	−0.14	130.979	−21.96	0.067	−6.44
M14	189	63.00	41	68.33	1.837	5.98	0.584	−0.03	126.899	−19.45	0.067	−7.52