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

Table 14

For K = 20, 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%	0.00407	0.07799	0.96028	0.01342	0.30048	0.86374
20%	65%	0.00436	0.08154	0.95852	0.00503	0.03857	0.98132
30%	55%	0.00469	0.08499	0.95721	0.00732	0.04650	0.96982
40%	45%	0.00578	0.10149	0.948802	0.00427	0.02725	0.98192
50%	35%	0.00707	0.127302	0.93491	0.00516	0.03518	0.98270
60%	25%	0.00944	0.15861	0.91822	0.00632	0.04150	0.97982
70%	15%	0.01535	0.23154	0.87938	0.00492	0.03154	0.98420
80%	05%	0.04379	0.47688	0.74082	0.00662	0.04481	0.97861