Table 7 ANOVA Post-hoc analysis (Dunn’s Test) results for additive regression, random subspace, linear regression, M5P, and SVM techniques during the validation phase in minimum air temperature.

From: Air temperature estimation and modeling using data driven techniques based on best subset regression model in Egypt

 

Sum Rank Diff

Z

Prob

Sig

“Actual” “Additive Regression"

1669

6.24027

6.5524E-9

1

“Actual” “Random Subspace"

2704

10.11005

7.47853E-23

1

“Actual” “Linear Regression"

6142.5

22.96635

1.51748E-115

1

“Actual” “M5P"

3286.5

12.28798

1.57654E-33

1

“Actual” “SVM"

−7055

26.37812

3.66447E-152

1

“Additive Regression” “Random Subspace"

1035

3.86979

0.00163

1

“Additive Regression” “Linear Regression"

4473.5

16.72608

1.26938E-61

1

“Additive Regression” “M5P"

1617.5

6.04771

2.20377E-8

1

“Additive Regression” “SVM"

−8724

32.61838

3.37606E-232

1

“Random Subspace” “Linear Regression"

3438.5

12.85629

1.18931E-36

1

“Random Subspace” “M5P"

582.5

2.17792

0.44118

0

“Random Subspace” “SVM"

−9759

36.48817

2.56222E-290

1

“Linear Regression” “M5P"

−2856

10.67837

1.92761E-25

1

“Linear Regression” “SVM"

−13197.5

49.34447

0

1

“M5P” “SVM"

−10341.5

38.6661

0

1

  1. Sig equals 1 indicates that the difference of the means is significant at the 0.05 level. Sig equals 0 indicates that the difference of the means is NOT significant at the 0.05 level.