A New Approach for Reconstruction of IMFs of Decomposition and Ensemble Model for Forecasting Crude Oil Prices

Table 6

values of Pearson correlation test for all samples and threshold values.

Data		Scenario I	Scenario II	Scenario III	Scenario IV
500	0.500	0.703	0.320	0.935	—
	0.900	0.853	0.518	0.454	0.890
	0.950	0.252	0.518	0.454	0.360
	0.990	0.252	<0.001	0.153	0.360
1000	0.500	0.673	0.240	0.619	—
	0.900	0.783	0.300	0.330	0.858
	0.950	0.783	0.329	0.217	0.938
	0.990	<0.001	<0.001	0.044	0.938
2000	0.500	0.624	0.192	0.452	0.509
	0.900	0.727	0.514	0.193	0.720
	0.950	0.727	0.514	0.193	0.528
	0.990	0.838	0.563	0.001	0.528
5000	0.500	<0.001	—	0.091	0.357
	0.900	0.131	0.163	0.101	0.625
	0.950	0.663	0.163	0.101	0.625
	0.990	0.747	0.298	0.001	0.003
10,000	0.500	0.039	—	0.173	0.175
	0.900	0.138	0.019	0.140	0.524
	0.950	0.369	0.083	0.140	0.524
	0.990	0.605	0.283	<0.001	0.006

The bold values in Table 6 are statistically significant at 5% and communicated that the two components are not independently distributed.