How to estimate and forecast FIGARCH model

[Hieu Nguyen Phi]

Sorry for this stupid question but I'm totally new here.
I'm currently writing my BA dissertation on estimating and forecasting with FIGARCH model by Chung (1999). Are there any softwares to do that? I've found on Eviews and Stata and R but it seems there is nothing at all.
Thank you so much.

 

[Alexandre Chretien]

Sorry for this stupid question but I'm totally new here.
I'm currently writing my BA dissertation on estimating and forecasting with FIGARCH model by Chung (1999). Are there any softwares to do that? I've found on Eviews and Stata and R but it seems there is nothing at all.
Thank you so much.

You are mostlikely going to have to program it by yourself. Even most typical variances of GARCH models have not been implemented in most languages.

 

[Hieu Nguyen Phi]

You are mostlikely going to have to program it by yourself. Even most typical variances of GARCH models have not been implemented in most languages.

Thanks for your reply.
I know nothing about programing. Please instruct me how to program to estimate a FIGARCH model?
Thanks.

 

[Alexandre Chretien]

If you know nothing about programming, I can't really help you.
Why don't you work with a simple Garch model?

 

[Hieu Nguyen Phi]

Actually, I learnt just a little bit about R.
S-Plus has a stronger ability to deal with FIGARCH but it has become commercial software. I suppose similar language like R can also do that. However, GarchOxFit function seems no longer in R library.
I found that FIGARCH is a totally new area and it might have some applications for my country. I found tens of studies about FIGARCH and none of them said about softwares or codes to run that model.
This site:Which software can help me to estimate the long memory GARCH? said Stata 14 can do that but I have no accounts for this site due to lack of institutional emails. So I guess I'm unable to contact that guy.
I found some Matlab code: https://www.kevinsheppard.com/images/9/95/MFE_Toolbox_Documentation.pdf but it seems to be for BBM FIGARCH (1996) rather than Chung FIGARCH (1999). Furthermore, my model contains some exogeneous variables in variances as dummies in order to consider various effects. I wonder whether it'll be possible if I change to use BBM FIGARCH and follow those codes.
I also found that best software to deal with FIGARCH is OxMetrics but it's actually expensive for a student like me.
I really hope that my thesis can eventually succeed.

 

[Alexandre Chretien]

I haven't look at the codes from the matlab toolbox, but usually it is the easier way to start. These two models might have a lot in common.

Do you know what are the differences between the 2 models and why you want to use one more than the other? You might be able to use the one provided in the paper you posted.

 

[Hieu Nguyen Phi]

FIGARCH of Chung is just a modified version of BBM in order to interprete more easily the differencing operation.
How do I start now? I feel so confused now.