A Hybrid Approach of Stepwise Regression, Logistic Regression, Support Vector Machine, and Decision Tree for Forecasting Fraudulent Financial Statements

Table 8

Selection of the research variables.

Variable classification	Variable code	Variable description and computation
Financial variables		Accounts receivables/total assets
		Gross profit/total assets
		Inventory/current assets
		Inventory/total assets
		Net profit after tax/total assets
		Net profit after tax/fixed assets
		Cash/total assets
		Log total assets
		Log total liabilities
		Interest protection multiples (debt service coverage ratio, times interest earned)
		Gross profit margin
		Operating expense ratio
		Debt ratio
		Inventory turnover
		Cash flow ratio
		Net profit ratio before tax
		Accounts payable turnover
		Revenue growth rate
		Debt/equity ratio
		Earnings before interest, taxes, depreciation, and amortization
		Current liabilities/total assets
		Total assets turnover
		Account receivable/last year accounts receivable >1.1
		Operation profit/last year operation profit >1.1
Nonfinancial variables		Shareholding ratio of the major shareholders
		Shareholding ratio of directors and supervisors
		Whether the chairman concurrently holds the position of CEO
		Board size
		Pledge ratio of shares of the directors and supervisors