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

Table 8

Performance of MLPNN and FTLRNN on testing data for Duffing time series.

MLPNN on testing data				FTLRNN on testing data
𝐾 (step)	MSE	NMSE	𝑟	MSE	NMSE	𝑟
1	0.08750	0.5666	0.66035	0.00113	0.00957	0.99545
5	0.07634	0.55568	0.68434	0.00202	0.01315	0.99356
10	0.08331	0.54275	0.67991	0.00335	0.02185	0.98910
20	0.08580	0.62975	0.63454	0.00632	0.04150	0.97982
50	0.10079	0.74545	0.54593	0.00927	0.06562	0.96663
100	0.11397	0.85011	0.40191	0.00117	0.07870	0.96000