Chinese Stock Index Futures Price Fluctuation Analysis and Prediction Based on Complementary Ensemble Empirical Mode Decomposition

Table 5

The prediction errors of different components.

	Daily price series		Five-minute price series
	Linear kernel function	RBF kernel function	Linear kernel function	RBF kernel function
Residue term
RMSE	0.007	0.008	0.006	0.009
MAPE	0.001	0.001	0.001	0.002
DS	69.837%	58.424%	21.678%	27.241%
Low-frequency term
RMSE	0.007	0.008	0.004	0.008
MAPE	0.087	0.092	0.042	0.258
DS	60.326%	55.434%	54.613%	55.254%
High-frequency term
RMSE	0.086	0.098	0.020	0.021
MAPE	4.806	4.725	1.146	1.149
DS	63.315%	62.500%	51.463%	51.854%