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

Table 7

Performance of MLPNN and FTLRNN on testing data for Mackey-Glass time series.

MLPNN on testing data				FTLRNN on testing data
𝐾 (step)	MSE	NMSE	𝑟	MSE	NMSE	𝑟
1	0.007	0.01	0.9933	0.000491	0.00919	0.99542
5	0.0028	0.051	0.9745	0.00227	0.04145	0.97911
10	0.0056	0.1008	0.9485	0.004475	0.07940	0.9596
20	0.0128	0.2159	0.8860	0.00944	0.15861	0.891822
50	0.0411	0.599	0.6406	0.024382	0.35655	0.80849
100	0.08180	0.9988	0.1971	0.05345	0.65267	0.62265