A Data-Characteristic-Driven Decomposition Ensemble Forecasting Research on the Demand of Space Science Payload Components

Table 7

Experimental results of the proposed model and comparison model.

Model	Performance evaluation
	Training set	Testing set
	MSE	MAPE	RMSE	SDAPE
ARIMA	2.0375	4.2755%	0.4858	4.2329	40%
GM(1,1)	2.5430	6.3494%	0.6975	6.2861	40%
ANN	3.8137	4.8912%	0.6155	4.8426	40%
MPSO-SVR	0.0001	6.7824%	0.7679	6.7147	0%
EMD-CC&CV-MPSO-SVR	0.2985	3.3381%	0.3414	3.3047	100%
EEMD-MPSO-SVR	0.0013	3.8326%	0.4497	3.7944	60%
EEMD-CC-MPSO-SVR	0.2887	2.4358%	0.2959	2.4115	100%
EEMD-CV-MPSO-SVR	0.1396	3.9024%	0.4460	3.8635	40%
EEMD-GS-SVR	0.1572	3.4780%	0.4151	3.4434	60%
EEMD-CC&CV-GA-SVR	0.0949	3.4233%	0.4385	3.3893	60%
EEMD-CC&CV-PSO-SVR	0.1096	4.7800%	0.4776	4.7322	80%
EEMD-CC&CV-MPSO-ANN	0.1026	4.2522%	0.4831	4.2098	60%
EEMD-CC&CV-MPSO-LSTM	0.1583	3.9122%	0.4609	3.8732	40%
EEMD-CC&CV-MPSO-SVR	0.0628	2.5215%	0.2774	2.4963	100%