Table 11 Estimated prediction accuracy (correlation between predicted and adjusted grain yield, averaged over 50 TRN-TST partitions, cycle 2012-2013, (W3), CV2
Models/EnvironmentsaCorrelation (SE)Change%bNumberc
Single environment
 0iFN0.559(0.036)
 2iBN0.448(0.045)
 5iBH0.630(0.035)
 5iBN0.356(0.060)
 5iFN0.307(0.042)
Pairs of environmentsInteraction ModelAcross-Env Model
 0iFNCor = 0.170.573(0.037)0.356(0.053)2.4%; 61.2%42; 50
 2iBN0.465(0.040)0.294(0.051)4.0%; 58.3%48; 50
 0iFNCor = 0.300.581(0.035)0.461(0.045)3.8%; 25.9%46; 50
 5iBH0.645(0.037)0.512(0.044)2.4%; 26.1%37; 50
 0iFNCor=-0.100.553(0.037)0.233(0.069)−1.2%; 136.7%2; 50
 5iBN0.347(0.043)0.058(0.052)−2.4%; 498.2%8; 50
 0iFNCor=-0.010.555(0.035)0.296(0.058)−0.8%; 87.5%7; 50
 5iFN0.302(0.042)0.087(0.060)−1.8%; 247.4%26; 50
 2iBNCor=-0.030.444(0.051)0.180(0.047)−0.8%; 146.8%27; 50
 5iBH0.633(0.037)0.337(0.052)0.4%; 87.6%13; 50
 2iBNCor = 0.120.446(0.048)0.321(0.047)−0.3%; 39.2%25; 50
 5iBN0.361(0.043)0.279(0.048)1.3%; 29.5%31; 49
 2iBNCor = 0.040.446(0.049)0.255(0.061)−0.4%; 74.6%31; 50
 5iFN0.307(0.045)0.142(0.057)−0.2%; 115.3%31; 50
 5iBHCor=-0.090.623(0.038)0.321(0.074)−1.1%; 94.1%3; 50
 5iBN0.345(0.045)0.037(0.058)−2.9%; 841.3%9; 50
 5iBHCor = 0.020.627(0.038)0.380(0.066)−0.5%; 65.1%24; 50
 5iFN0.305(0.043)0.106(0.059)−0.6%; 188.3%29; 50
 5iBNCor = 0.550.609(0.050)0.570(0.055)71.2%; 7.0%50; 49
 5iFN0.582(0.037)0.542(0.043)89.3%; 7.3%50; 49
All environments
 0iFN0.575(0.034)0.301(0.057)2.7%; 90.9%50; 50
 2iBN0.466(0.043)0.217(0.052)4.1%; 114.7%43; 50
 5iBH0.629(0.035)0.281(0.046)−0.2%; 124.3%24; 50
 5iBN0.402(0.055)0.243(0.054)12.9%; 65.3%50; 50
 5iFN0.376(0.041)0.244(0.055)22.3%; 53.8%50; 49
  • TRN-TST, training-testing.

  • a Environments 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.

  • b Change 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).

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