Crude Oil Price Forecasting Based on Hybridizing Wavelet Multiple Linear Regression Model, Particle Swarm Optimization Techniques, and Principal Component Analysis

Table 2

Forecasting performance indices of MLR and WLR.

Model Input	Lag	MLR			WMLR
		MAE	RMSE	𝐷stat	MAE	RMSE	𝐷stat
M1	1	0.6948	0.9514	0.4788	0.6660	0.9001	0.5198
M2	1, 2	0.6972	0.9517	0.4781	0.6448	0.8842	0.5003
M3	1, 2, 3	0.6985	0.9545	0.4816	0.5345	0.7505	0.5797
M4	1, 2, 3, 4	0.6979	0.9550	0.4753	0.4834	0.6572	0.6722
M5	1, 2, 3, 4, 5	0.6976	0.9545	0.4878	0.5770	0.8046	0.5734
M6	1, 2, 3, 4, 5, 6	0.6969	0.9450	0.4850	0.5385	0.7389	0.6444