The Efficacy of Predictive Methods in Financial Statement Fraud

Table 6

Descriptive statistics of the performance measures obtained from 1000 iterations segmented by different rate of fraudulent companies, and for training sample size of 800 and test sample size of 270.

		1% Fraud		3% Fraud		5% Fraud		7% Fraud		10% Fraud
		Prec.	recall	Prec.	recall	Prec.	recall	Prec.	recall	Prec.	recall
MFFNN	Avg.	0.975	0.969	0.974	0.966	0.972	0.965	0.972	0.964	0.973	0.962
	SD	0.015	0.017	0.016	0.018	0.016	0.019	0.016	0.019	0.017	0.019
	Min	0.907	0.897	0.917	0.881	0.916	0.878	0.898	0.893	0.907	0.886
	Max	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000
PNN	Avg.	0.975	0.726	0.974	0.720	0.974	0.720	0.975	0.718	0.976	0.718
	SD	0.015	0.035	0.016	0.034	0.016	0.034	0.016	0.034	0.015	0.035
	Min	0.916	0.623	0.906	0.630	0.906	0.614	0.879	0.631	0.916	0.578
	Max	1.000	0.843	1.000	0.846	1.000	0.822	1.000	0.835	1.000	0.855
SVM	Avg.	0.916	0.921	0.909	0.916	0.905	0.914	0.906	0.914	0.907	0.916
	SD	0.028	0.027	0.028	0.026	0.029	0.028	0.030	0.026	0.028	0.026
	Min.	0.796	0.828	0.796	0.811	0.813	0.800	0.806	0.833	0.796	0.840
	Max	0.981	0.990	0.991	1.000	0.972	0.990	0.981	0.990	0.972	0.980
MLM	Avg.	0.894	0.897	0.892	0.894	0.890	0.893	0.892	0.895	0.894	0.898
	SD	0.032	0.027	0.032	0.027	0.032	0.028	0.032	0.027	0.029	0.027
	Min	0.759	0.804	0.787	0.809	0.785	0.787	0.787	0.805	0.806	0.786
	Max	0.981	0.980	0.991	0.980	0.972	0.980	0.981	0.980	0.981	0.970
DA	Avg.	0.912	0.870	0.908	0.877	0.906	0.884	0.907	0.892	0.905	0.902
	SD	0.029	0.029	0.029	0.028	0.030	0.028	0.029	0.027	0.029	0.026
	Min	0.806	0.784	0.815	0.800	0.804	0.771	0.806	0.800	0.796	0.814
	Max	0.991	0.971	0.991	0.961	0.981	0.979	0.991	0.980	0.991	0.980