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

Table 9

Predictive results of six models based on four evaluation criteria.

Model	Type-I accuracy (%)	Type-I error (%)	Type-II accuracy (%)	Type-II error (%)
Logistic regression	80.33	19.67	85.71	14.29
Neural networks	91.67	8.33	72.86	27.14
MCLOC	89.33	10.67	68.57	31.43
MCQOC	90.67	9.33	70.71	29.29
SVM	94.67	5.33	77.14	22.86
KFS-MCLOC	91.33	8.67	98.57	1.43