Table 4 Multi-environment prediction performance of seven kernels with a ordinal response variable under a ten-fold cross-validation with data Data_Wheat_2019.RData.

From: A guide for kernel generalized regression methods for genomic-enabled prediction

Kernel

PCCC

SE_PCCC

Kappa

SE_Kappa

Time

Linear

0.671

0.025

0.454

0.044

62.72

Polynomial

0.680

0.026

0.468

0.049

69.98

Sigmoid

0.664

0.025

0.443

0.045

71.25

Gaussian

0.689

0.028

0.483

0.054

72.1

AK1

0.664

0.027

0.441

0.050

79.83

AKL

0.673

0.025

0.450

0.048

65.6

Exponential

0.596

0.035

0.281

0.075

72.76

  1. PCCC denotes the proportion of cases correctly classified (SE_PCCC is the standard error of PCCC) and Kappa is average Kappa’s coefficient (SE_Kappa is the standard error of Kappa). Time indicates the implementation time in seconds.
Back to article page