Improving Forecasts of the EGARCH Model Using Artificial Neural Network and Fuzzy Inference System

Table 9

The EGARCH-ANN (2, 4, 0, 0) model fitted to the given data.

Parameters	Coefficients	Std. errors	t-stats
K	−2.5633	0.5817	−4.4067
ARCH1	0.1988	0.0744	2.6727
ARCH2	−0.2074	0.0777	−2.6704
GARCH1	1.1701	0.1324	8.8384
GARCH2	−1.1034	0.1666	−6.6238
GARCH3	0.8452	0.1675	5.0462
GARCH4	−0.4982	0.1543	−3.2291
Leverage1	−0.1534	0.0423	−3.6284
Leverage2	0.2297	0.0457	5.0208
Dof	15.002	7.1161	2.1082

With addition information like log likelihood: 608, Akaike information criterion: −1196, Bayesian information criterion: −1149, and Hannan–Quinn information criterion: −1178.