Credit Risk Assessment for Small and Microsized Enterprises Using Kernel Feature Selection-Based Multiple Criteria Linear Optimization Classifier: Evidence from China

Table 8

Predictive results of six models based on five evaluation criteria.

Model	TA (%)	score	AUC	KS score	MCC
Logistic regression	82.05	0.75	0.82	0.61	0.63
Neural networks	85.68	0.77	0.83	0.72	0.67
MCLOC	82.73	0.72	0.79	0.75	0.6
MCQOC	84.32	0.74	0.8	0.68	0.63
SVM	89.09	0.82	0.87	0.8	0.74
KFS-MCLOC	93.63	0.91	0.93	0.85	0.87