Forecasting Crude Oil Consumption in China Using a Grey Prediction Model with an Optimal Fractional-Order Accumulating Operator

Table 2

The simulation/prediction values and MAPE of the models for China’s crude oil consumption.

(a)

Year	Actual value	FSIGM	SIGM	GM	DGM	NDGM
		Simulation value	Simulation value	Simulation value	Simulation value	Simulation value
In-sample
2002	2.2544	2.2544	2.2544	2.2544	2.2544	2.2544
2003	2.4922	2.4889	2.5374	2.6229	2.6235	2.5374
2004	2.8749	2.8749	2.7897	2.7941	2.7945	2.7897
2005	3.0086	3.0305	3.0204	2.9764	2.9766	3.0204
2006	3.2245	3.2019	3.2314	3.1706	3.1707	3.2313
2007	3.4032	3.3868	3.4243	3.3774	3.3774	3.4243
2008	3.5498	3.5848	3.6007	3.5978	3.5975	3.6007
2009	3.8129	3.7962	3.7620	3.8325	3.8320	3.7260
MAPE (in)		0.4949%	1.2526%	1.9179%	1.9145%	1.2525%

(b)

		Prediction value	Prediction value	Prediction value	Prediction value	Prediction value
Out-of-sample
2010	4.2875	4.0215	3.9094	4.0826	4.0818	3.9095
2011	4.3966	4.2612	4.0443	4.3490	4.3479	4.0444
2012	4.6679	4.5160	4.1676	4.6327	4.6313	4.1678
2013	4.8652	4.7869	4.2804	4.9350	4.9332	4.2806
2014	5.1547	5.0746	4.3835	5.2569	5.2548	4.3837
MAPE (out)		3.1401%	10.9054%	2.0069%	9.1817%	10.9031%