Table 6. Estimated prediction accuracy: correlation between predicted and adjusted grain yield, averaged over 50 TRN-TST partitions), cycle 2010−2011 (W1), CV1.

Model/Environmentsa		Correlation		Change%b	Numberc
Single environment
0iBN		0.530(0.039)		−	−
2iBN		0.629(0.038)		−	−
5iBNZ		0.472(0.054)		−	−
5iFN		0.486(0.050)		−	−
Pairs of environments		Interaction Model	Across-Env Model
0iBN	Cor = 0.53	0.529(0.038)	0.523(0.042)	−0.3%; 1.1%	21; 33
2iBN		0.619(0.039)	0.593(0.041)	−1.6%; 4.3%	12; 49
0iBN	Cor = 0.25	0.527(0.041)	0.467(0.045)	−0.6%; 12.8%	19; 50
5iBNZ		0.468(0.053)	0.375(0.052)	−1%; 24.8%	8; 49
0iBN	Cor = 0.26	0.534(0.038)	0.494(0.044)	0.7%; 8%	40; 48
5iFN		0.485(0.050)	0.431(0.050)	−0.2%; 12.5%	26; 49
2iBN	Cor = 0.34	0.633(0.039)	0.576(0.045)	0.8%; 10%	37; 50
5iBNZ		0.477(0.050)	0.428(0.047)	0.9%; 11.2%	29; 46
2iBN	Cor = 0.33	0.625(0.041)	0.555(0.044)	−0.6%; 12.5%	18; 50
5iFN		0.479(0.051)	0.422(0.056)	−1.4%; 13.5%	15; 49
5iBNZ	Cor = 0.22	0.466(0.053)	0.406(0.056)	−1.3%; 14.8%	15; 50
5iFN		0.483(0.051)	0.448(0.053)	−0.6%; 8%	18; 48
All environments
0iBN		0.530(0.041)	0.483(0.048)	0%; 9.7%	24; 49
2iBN		0.625(0.042)	0.558(0.046)	−0.5%; 12.1%	19; 50
5iBNZ		0.462(0.050)	0.366(0.051)	−2.2%; 26.3%	13; 50
5iFN		0.470(0.049)	0.394(0.056)	−3.3%; 19.2%	9; 50

TRN-TST, training-testing.

^aEnvironments are described by a sequence of codes: 0i, 2i, and 5i denote the number of irrigation cycles; B/F denotes whether the planting system was “bed” (B) or “flat” (F); N/H denotes whether planting date was normal (N) or late (H, simulating heat); Z indicates no tillage.

^bChange in prediction accuracy of the M×E model relative to the prediction accuracy of the single-environment (before semicolon) and relative to the prediction accuracy of the across-environment model (after semicolon).

^cNumber of partitions (of 50) for which the M×E model gave greater accuracy than the single-environment (before semicolon) and the across-environment model (after semicolon).