Skip to main content

Table 31 Discrimination performance of different feature sets

From: Clues from networks: quantifying relational risk for credit risk evaluation of SMEs

Model

Metric

BF + CNAP

BF + CNLD

BF + DNAP

BF + DNLD

BF + SNAP

BF + SNLD

LR

AUC

0.725

0.731

0.732

0.736

0.727

0.728

KS

0.421

0.431

0.427

0.443

0.420

0.422

H

0.271

0.283

0.271

0.288

0.270

0.273

RF

AUC

0.739

0.747

0.739

0.749

0.738

0.745

KS

0.448

0.467

0.442

0.467

0.446

0.463

H

0.288

0.303

0.292

0.305

0.283

0.299

XGB

AUC

0.718

0.717

0.726

0.739

0.719

0.723

KS

0.420

0.420

0.440

0.449

0.420

0.431

H

0.249

0.252

0.252

0.292

0.252

0.257

  1. Each metric is measured in mean of 50 estimates obtained by repeating the outer ten-fold cross-validation procedure five times