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

Table 15

For K = 50, 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%	0.01082	0.19456	0.89823	0.01290	0.79033	0.71021
20%	65%	0.01169	0.20374	0.89309	0.01027	0.08318	0.95984
30%	55%	0.01316	0.22115	0.88438	0.00658	0.04203	0.97902
40%	45%	0.01529	0.24709	0.87032	0.00763	0.05034	0.97470
50%	35%	0.01816	0.29049	0.84538	0.00781	0.05426	0.97271
60%	25%	0.024382	0.35655	0.80849	0.00927	0.06562	0.96663
70%	15%	0.03829	0.48826	0.73543	0.00743	0.05020	0.97567
80%	05%	0.1056	0.87428	0.52529	0.00953	0.06451	0.96811