Novel FTLRNN with Gamma Memory for Short-Term and Long-Term Predictions of Chaotic Time Series

Table 11

For K = 1, training and testing samples variation for FTLRNN on testing data set.

Time series		Mackey-Glass time series			Duffing time series
Training exemplars	Testing exemplars	MSE	NMSE	𝑟	MSE	NMSE	𝑟
10%	75%	0.000315	0.00627	0.99707	0.00230	0.03972	0.98015
20%	65%	0.000247	0.004854	0.99757	0.00101	0.00774	0.99618
30%	55%	0.000405	0.007732	0.99628	0.00104	0.00707	0.99671
40%	45%	0.000371	0.00691	0.99663	0.00103	0.00655	0.99692
50%	35%	0.000371	0.007283	0.9964	0.00087	0.00555	0.99733
60%	25%	0.000491	0.00919	0.99542	0.00113	0.00957	0.99545
70%	15%	0.00078	0.01353	0.9933	0.00091	0.00609	0.99712
80%	05%	0.002217	0.03312	0.98337	0.00484	0.03271	0.98827