Table 15 Results of the Wilcoxon rank sum test for the three models.

From: Example dependent cost sensitive learning based selective deep ensemble model for customer credit scoring

Comparison

\(T = {\text{min}}\left( {R^{ + } ,R^{ - } } \right)\)

CV

p-value

Hypothesis

ECS-SDE VS ECSTabNet + SRCGMDH

min (192.0, 18.0) = 18.0

52

0.000

Reject

ECS-SDE VS TabNet + ECSGMDH

min (196.5, 13.5) = 13.5

52

0.000

Reject

ECSTabNet + SRCGMDH VS TabNet + ECSGMDH

min (132.0, 78.0) = 78.0

52

0.287

Accept