Page 1 of 1
GJR-GARCH(1,1) & TGARCH(1,1)
Posted: Mon Feb 24, 2014 4:57 am
by ecofin
hi everybody, I have seen thant there is diference between GJR-GARCH(1,1) and TGARCH(1,1) when I estimate the two models with Mathlab, my question is: EViews estimate GJR-GARCH(1,1) or TGARCH(1,1)?.
Reference: Financial modeling Under Non-Gaussian Distribution, Eric Jondeau, Ser-Huang Poon and Michael Rockinger, Springer 2007, pp 95 and pp 100.
Re: GJR-GARCH(1,1) & TGARCH(1,1)
Posted: Mon Feb 24, 2014 12:13 pm
by trubador
Re: GJR-GARCH(1,1) & TGARCH(1,1)
Posted: Thu Feb 27, 2014 11:43 am
by ecofin
thanks for your response, So there is not a way to estimate original TGARCH(1,1) in EViews 8.
Re: GJR-GARCH(1,1) & TGARCH(1,1)
Posted: Thu Feb 27, 2014 2:07 pm
by trubador
No, not built-in. But you can always build such customized models via LogL object.
Re: GJR-GARCH(1,1) & TGARCH(1,1)
Posted: Mon Mar 03, 2014 12:57 pm
by ecofin
but i don't know where to start, I'm not an expert on Eviews programming, Can you help me for the code to customized TGARCH(1,1), and thanks for your help and for your reply.
Re: GJR-GARCH(1,1) & TGARCH(1,1)
Posted: Mon Mar 03, 2014 3:31 pm
by trubador
You can start by examining the customized GARCH examples written via LogL object: "C:\Program Files\EViews 8\Example Files\Sample Programs\logl". And you can always search forum for similar examples...
Re: GJR-GARCH(1,1) & TGARCH(1,1)
Posted: Tue Mar 04, 2014 12:40 pm
by ecofin
thanks for your help and for your reply.
just i can estimate the original TARCH model with PARCH model when i fix the power=1 but i loss the threshod order :? normally it should leave this option
