Hybrid Time-Series Prediction Method Based on Entropy Fusion Feature

Table 4

Prediction evaluation at different iteration times.

Index	Algorithm	epoch10	epoch15	epoch20	epoch25	epoch50	epoch75	epoch100
RMSE	EEMD-ELSTM	1931.74	1165.97	909.42	841.73	575.16	550.01	556.59
	EEMD-LSTM	1951.56	1634.48	1184.47	1067.86	651.13	609.33	630.51
	ELSTM	5438.14	5367.18	5295.41	4956.89	4632.53	4376.05	4533.58
	LSTM	5582.33	5582.28	5301.50	5009.36	4739.33	4497.84	4634.36
MAE	EEMD-ELSTM	1650.04	942.45	702.52	640.11	449.06	411.75	439.26
	EEMD-LSTM	1654.95	1392.24	962.65	863.52	528.75	480.41	497.95
	ELSTM	4598.59	4565.94	4541.52	4372.06	4222.25	3909.26	4286.08
	LSTM	4662.28	4660.97	4542.76	4433.95	4366.71	4189.13	4325.04
MAPE (%)	EEMD-ELSTM	4.77	2.77	2.03	1.85	1.29	1.22	1.31
	EEMD-LSTM	4.81	4.01	2.81	2.55	1.55	1.54	1.55
	ELSTM	12.66	12.60	12.51	12.14	12.01	11.80	11.92
	LSTM	12.75	12.72	12.53	12.33	12.45	11.89	12.32
	EEMD-ELSTM	0.96	0.99	0.99	0.99	0.99	0.99	0.99
	EEMD-LSTM	0.96	0.97	0.98	0.99	0.99	0.99	0.99
	ELSTM	0.91	0.91	0.92	0.94	0.96	0.97	0.97
	LSTM	0.88	0.88	0.91	0.94	0.96	0.97	0.97