I did GARCH(1,1) and result show that only Alpha or Beta is significant. Can I interpret them separately? If yes, what is the alpha and beta tell us in practical and what is the difference?
Last question, I knew that alpha+beta that close to 1 is mean a persistance of volatility but in my case, only 1 variable is significant. So, this fact is still usable?
Thanks in advance for the answer. I know this question is very simple. But I searched for a lot of sources and research. but I still don't get a satisflying answer.
It is not as simple as you may think. The usual tests do not work in that case. Have a look at the paper by Francq, Christian, and Jean-Michel Zakoïan. “Testing the Nullity of GARCH Coefficients: Correction of the Standard Tests and Relative Efficiency Comparisons.” Journal of the American Statistical Association, vol. 104, no. 485, 2009, pp. 313–324. I don't know if the software you have used (EViews, apparently) has made the necessary corrections. Good luck.
The GARCH term is the coefficient of GARCH (-1) and is positive and statistically significant. As the GARCH coefficient value is higher than the ARCH coefficient value, we can conclude that the volatility is highly persistent and clustering1
Chuck A Arize Thank foryour answer Chuck. What if ARCH term? Can I say that the past volatility is affect to current one but it not persisting?
Did you have a look in EViews' manual to see what they do exactly? Do they follow the Francq-Zakoïan recommendations? Otherwise, Chuck A Arize 's answer is only approximate.
- Badges
- Science topic
- Similar topics
- Correction