Hi guys,
I am new to eviews and coding in general.
I am doing a project testing the volatility forecast abilities of various models, GARCH(1,1) included. Things have been going well but now I am stuck and have been for over a week.
I have been trying to develop the code for a rolling forecast of a GARCH(1,1) model without any success.
My sample size is 1250. I am using the first 1000 observations to estimate the GARCH coefficients, do the 1 day ahead forecast of the conditional variance, then drop the first observation and add the 1001st to re estimate the coefficients, do the next 1 day ahead forecast and so on….
Can anyone help me with this as it is really getting frustrating.
Thanks a lot.
Ina
Rolling forecast GARCH
Moderators: EViews Gareth, EViews Jason, EViews Moderator, EViews Matt
-
- Posts: 2
- Joined: Sun Jun 28, 2009 7:41 am
Re: Rolling forecast GARCH
Thanks.
I have looked at that example before and customized it but unfortunately it didn’t seem to work or better: I didnt get it to work properly.
Regards,
Ina
I have looked at that example before and customized it but unfortunately it didn’t seem to work or better: I didnt get it to work properly.
Regards,
Ina
Re: Rolling forecast GARCH
I am attempting to do a GARCH in the same fasion, but stuck on the code. Can anyone assist please?
Code: Select all
!window = 200
!step = 1
!length = @obsrange
!nrolls = @round((!length-!window)/!step)
matrix(1,!nrolls) results
!j=0
for !i = 1 to !length-!window+1-!step step !step
!j=!j+1
smpl @first+!i-1 @first+!i+!window-2
equation garch_n
garch_n.ARCH(1,1) returns
rowplace(results,garch_n@ARCH(1,1),!j)
next
show results
Who is online
Users browsing this forum: No registered users and 41 guests