Chinese Currency Exchange Rates Forecasting with EMD-Based Neural Network

Table 4

Robust tests for NMSE comparisons with different subsamples.

	1-day	5-day	10-day	20-day	30-day
Panel A: MLP model
MLP(3)	0.0175	0.0740	0.2010	0.3638	0.5954
MLP(5)	0.0201	0.0699	0.2082	0.3663	0.5158
MLP(5,3)	0.0212	0.0761	0.2117	0.3614	0.5561
MLP(6,4)	0.0221	0.0694	0.2066	0.3586	0.5746
Panel B: EMD-MLP model
EMD(-1)-MLP(3)		0.0705		0.3827	0.5885
EMD(-1)-MLP(5,3)		0.0713
EMD(-2)-MLP(3)	0.0192			0.3568
EMD(-2)-MLP(5,3)	0.0222				0.5417
EMD(-3)-MLP(3)	0.0423
EMD(-3)-MLP(5,3)	0.0442
Panel C: EMD-MLP model
EMD(0)-MLP(3)
EMD(0)-MLP(5,3)
EMD(-1)-MLP(3)
EMD(-1)-MLP(5,3)
EMD(-2)-MLP(3)	0.0191
EMD(-2)-MLP(5,3)	0.0201

Consider the CNY from January 2, 2006, to December 21, 2015, with a total of 2584 observations. For training the neural network models, three-fourths of the observations are randomly assigned to the training dataset and the remainder is used as the testing dataset. This table compares the forecasting performance, in terms of the NMSE, for the MLP, EMD-MLP, and EMD-MLP models. We report the NMSE as percentage for l-day ahead predictions where and 30. The DM test [38] is used to compare the forecast accuracy of EMD-MLP (EMD-MLP) model and the corresponding MLP model. , , and denote statistical significance at 1%, 5%, and 10%, respectively. For each length of prediction, we mark the minimum NMSE as bold.