Table 2 Relative performance of different ensemble methods using all level 1 inference methods’ results (i.e., regardless of kurtosis) as ensemble inputs and the same models while only using level 1 inference methods’ results with positive kurtosis, marked by plus signs (corresponding to positive kurtosis)

1.11 ± 0.45

0.99 ± 0.21

0.92 ± 0.34

1.33 ± 0.56

1.25 ± 0.44

1.13 ± 0.48

1.12 ± 0.43

0.95 ± 0.35

1.31 ± 0.54

1.24 ± 0.46

0.25 ± 0.06

0.4 ± 0.47

0.2 ± 0.11

0.2 ± 0.14

0.26 ± 0.19

0.71 ± 0.23

0.63 ± 0.19

0.54 ± 0.08

0.55 ± 0.15

0.57 ± 0.21

0.83 ± 0.23

0.71 ± 0.17

0.65 ± 0.15

0.67 ± 0.16

0.68 ± 0.26

1.33 ± 0.58

1.09 ± 0.26

1.19 ± 0.5

1.14 ± 0.48

1.13 ± 0.35

0.21 ± 0.09

0.29 ± 0.14

0.26 ± 0.09

0.25 ± 0.15

0.26 ± 0.22

0.47 ± 0.2

0.51 ± 0.19

0.43 ± 0.12

0.86 ± 0.6

0.47 ± 0.25

0.59 ± 0.24

0.45 ± 0.22

0.51 ± 0.2

0.39 ± 0.34

0.57 ± 0.41

0.58 ± 0.23

0.68 ± 0.54

0.71 ± 0.47

0.6 ± 0.46

0.65 ± 0.31

1.12 ± 0.56

0.9 ± 0.21

0.99 ± 0.29

1.73 ± 1.03

1.09 ± 0.38

1.1 ± 0.51

0.97 ± 0.42

1.04 ± 0.37

1.73 ± 1.1

1.08 ± 0.34

1.12 ± 0.5

1.13 ± 0.76

0.73 ± 0.26

1.27 ± 0.65

0.95 ± 0.46

1.07 ± 0.43

1.11 ± 0.65

0.82 ± 0.27

1.3 ± 0.7

1.03 ± 0.49

0.65 ± 0.34

0.56 ± 0.16

0.61 ± 0.28

1.35 ± 0.86

0.66 ± 0.24

0.6 ± 0.25

0.58 ± 0.23

0.78 ± 0.63

1.31 ± 0.86

0.67 ± 0.26

0.88 ± 0.23

0.88 ± 0.2

0.88 ± 0.24

1 ± 0

0.81 ± 0.22

0.93 ± 0.26

1.11 ± 0.58

1.08 ± 0.34

1.03 ± 0.49

0.95 ± 0.44

0.95 ± 0.42

0.88 ± 0.52

0.78 ± 0.39

0.69 ± 0.16

0.61 ± 0.22