EViews.com

dear members
if using eviews we capture the leverage effects in asset series, is it necessary to model FIGARCH for modelling long memory or FIEGARCH is enough?
regards
hojat

Since a FIGARCH model does not account for leverage effect, you should begin with a more generalized version, FIEGARCH that is. The difference between FIGARCH and FIEGARCH is analogous to the difference between EGARCH and GARCH. If the coefficient of the leverage effect turns out to be insignificant, then you can continue your analysis with FIGARCH.

thanks for your kind reply.i am using s-plus to model fiegarch .i read the zivot book.they suggest to use the absoult of returns series.with this i get 0.244 for fraction and using log returns series i get 0.44 for fraction.in both case it is sign of long memory assuming the fiegarch(1,1) is the fitted model.as you know in eviews we have SBIC AIC to determine the order of the model.and we can use the ARCH-LM test to test the adequacy of the model.but unfortunatly s-plus require excellancy in programming particularly S language.i dont know how to apply these in eviews.any suggestion in this regrda will be appreciated.thanks.

dear members
if using eviews we capture the leverage effects in asset series, is it necessary to model FIGARCH for modelling long memory or FIEGARCH is enough?
regards
hojat

How can I run FIEGARCH on Eviews 8? or it is not availlable?

How can I run FIEGARCH on Eviews 8? or it is not availlable?