Table 4. Maize data.
Correlation | MSPE | ||||||
---|---|---|---|---|---|---|---|
BMTME | Environment–Trait | Mean | SE | Ranka | Mean | SE | Rank |
E1-Yield | 0.28 | 0.07 | 3 | 0.74 | 0.08 | 3.00 | |
E2-Yield | 0.40 | 0.09 | 1.5 | 0.39 | 0.06 | 2.50 | |
E3-Yield | 0.37 | 0.08 | 2.5 | 0.02 | 0.01 | 1.50 | |
E1-ASI | 0.39 | 0.08 | 2 | 0.37 | 0.03 | 1.50 | |
Unstructured | E2-ASI | 0.46 | 0.06 | 2.5 | 1.26 | 0.35 | 3.00 |
E3-ASI | 0.42 | 0.05 | 2 | 0.01 | 0.00 | 1.50 | |
E1-PH | 0.37 | 0.06 | 2 | 0.86 | 0.07 | 3.00 | |
E2-PH | 0.26 | 0.08 | 3 | 0.48 | 0.07 | 3.00 | |
E3-PH | 0.44 | 0.07 | 2.5 | 0.02 | 0.00 | 2.00 | |
Average | 0.37 | 0.07 | 2.33 | 0.46 | 0.08 | 2.33 | |
E1-Yield | 0.30 | 0.07 | 1.5 | 0.73 | 0.07 | 2.00 | |
E2-Yield | 0.40 | 0.08 | 1.5 | 0.36 | 0.03 | 1.00 | |
E3-Yield | 0.37 | 0.05 | 2.5 | 0.85 | 0.07 | 3.00 | |
E1-ASI | 0.40 | 0.09 | 1 | 0.39 | 0.06 | 3.00 | |
Diagonal | E2-ASI | 0.46 | 0.06 | 2.5 | 1.25 | 0.35 | 2.00 |
E3-ASI | 0.27 | 0.08 | 3 | 0.48 | 0.07 | 3.00 | |
E1-PH | 0.36 | 0.07 | 3 | 0.02 | 0.01 | 1.00 | |
E2-PH | 0.41 | 0.06 | 1 | 0.01 | 0.00 | 1.00 | |
E3-PH | 0.44 | 0.06 | 2.5 | 0.02 | 0.00 | 2.00 | |
Average | 0.38 | 0.07 | 2.06 | 0.46 | 0.08 | 2.00 | |
E1-Yield | 0.30 | 0.07 | 1.5 | 0.72 | 0.07 | 1.00 | |
E2-Yield | 0.39 | 0.09 | 3 | 0.39 | 0.06 | 2.50 | |
E3-Yield | 0.38 | 0.08 | 1 | 0.02 | 0.01 | 1.50 | |
E1-ASI | 0.38 | 0.08 | 3 | 0.37 | 0.03 | 1.50 | |
Standard | E2-ASI | 0.48 | 0.06 | 1 | 1.24 | 0.35 | 1.00 |
E3-ASI | 0.43 | 0.05 | 1 | 0.01 | 0.00 | 1.50 | |
E1-PH | 0.39 | 0.06 | 1 | 0.84 | 0.06 | 2.00 | |
E2-PH | 0.27 | 0.08 | 2 | 0.47 | 0.07 | 2.00 | |
E3-PH | 0.45 | 0.07 | 1 | 0.02 | 0.00 | 2.00 | |
Average | 0.39 | 0.07 | 1.61 | 0.45 | 0.07 | 1.67 |
Mean and SE of the estimated correlations and MSPE from the 10-fold cross-validation CV1. The BMTME model was fitted using unstructured, diagonal, and standard variance–covariance matrices. Environment (E1, E2, E3)–trait (Yield, ASI, PH) combination.
Since three BMTME models are fitted (unstructured, diagonal, and standard) the values of the ranks ranged from 1 to 3, and the lower the values, the better the prediction accuracy. For ties, we assigned the average of the ranks that would have been assigned had there been no ties.