Table 2 WHEAT1 data set. Average Pearson’s correlations between observed and predictive values (and their standard deviation in parentheses) for three methods for a GE multi-environment model for 50 random partitions with 70% of the lines in the training set and 30% of the lines in the testing set. Methods GB, GK, and AK are the GBLUP, Gaussian Kernel and Arc-Cosine Kernel, respectively, with six levels (layers). The best predictive model for each environment (E1-E4) is in boldface
GE multi-environment model
E10.422 (0.05)0.494+ (0.06)60.482 (0.06)
E20.537 (0.04)0.582+ (0.04)60.581 (0.04)
E30.441 (0.05)0.497+ (0.06)60.494 (0.05)
E40.485 (0.05)0.558+ (0.04)60.551 (0.05)
  • + Significant at the 0.05 probability level of the t-test for the hypothesis that the average of the correlation of kernel AK is superior to the mean of the correlation of kernel GB.