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

Table 11

Feature selection results by KFS-MCLOC.

Variable	Variable name	Category	Positive/negative	Weight
	Quick ratio	Enterprises’ financial information	Negative	−2.04%
	Operating net cash flow	Enterprises’ financial information	Negative	−19.76%
	Actual controller’s management experience	Actual controller’s information	Negative	−2.52%
	Whether the actual controller has any record of tax default and illegal behaviour	Actual controller’s information	Positive	7.95%
	Whether the wages of employees in the past 12 months can be paid on time	Enterprises’ behaviour information	Negative	−7.21%
	Abnormal times of water, electricity, and tax in the past 12 months	Enterprises’ behaviour information	Positive	21.42%
	Overdue days of enterprises’ loans	Enterprises’ behaviour information	Positive	10.14%
	Mortgage and pledge status	Enterprises’ behaviour information	Negative	−13.09%
	Whether the guarantee enterprise is abnormal	Enterprises’ behaviour information	Positive	10.06%
	Number of administrative punishments	Enterprises’ supervision information	Positive	5.81%