Detecting Falsified Financial Statements Using a Hybrid SM-UTADIS Approach : Empirical Analysis of Listed Traditional Chinese Medicine Companies in China

Table 4

Forecasting results with the trained SM-UTADIS model (testing dataset).

Considered company	Year	Actual class	Utility value	Estimated class
A1	2002	C1	0.349590	C1
	2004	C1	0.349710	C1
	2010	C1	0.358811	C1
	2012	C1	0.386387	C1
A2	2005	C1	0.350307	C1
	2006	C1	0.380931	C1
	2008	C1	0.357234	C1
	2014	C1	0.312229	C2
A3	2004	C1	0.352612	C1
	2009	C1	0.350430	C1
	2011	C1	0.351601	C1
	2014	C1	0.365781	C1
	Utility threshold t		0.349489
A1	2006	C2	0.316715	C2
	2008	C2	0.353189	C1
A2	2010	C2	0.325583	C2
	2012	C2	0.339748	C2
A3	2007	C2	0.345021	C2
	2013	C2	0.050622	C2