Rolling GARCH with forecasting

[EViews Gareth]

Still haven't told us what you did though.

[stasibab]

Ok I will try to be as self-explanatory as I can.

I built a GARCH (1,1) model for the series 'sample' as you can see from the specification of the equation in the file. So I put in the mean equation ' sample c' and then in GARCH 1 and ARCH 1 and threshold 0. I used the first 1800 observations for the insample. Then I forecasted using the whole sample (2295) in order to get an outofsample forecast. In the methods I used static method. And in the series name I put sef and volf which is the outofsample forecast for standard errors and volatility. In this forecast my last observation appears as N.A.

[EViews Gareth]

I just did those exact steps and got values for the last period. Perhaps you need to update your copy of EViews.

[stasibab]

Dear Gareth,

I have an updated copy of eviews 7. I am interested in forecasting with GARCH the conditional volatility of my series (sample in the Eviews file attached).I estimated GARCH with the first 1800 observations of my dataset. I click forecast, sample c for the mean equation, click static and I named the conditional volatility forecasts for the whole period volf. However, on the results I got NA for the last forecast (observation 2295).
I replicated this procedure in different pcs with Eviews and I got the same error. Is it possible that all this is a bug of Eviews? Or I am missing something? According to the handbookl of Eviews, on cell 2295 of volf should be the conditional volatility forecast of my series for day 2295.I am working at a university which use multiple legitime copies of EVIEWS 7.

[EViews Gareth]

What's the build date of the copy of EViews (Help->About EViews)?

[stasibab]

Hi again,

The build version the university uses is January 7 2010.

I also tried the same with Eviews 5 and I did not have any problems. So probably there is something with this version. I do not know if you have any other suggestion.

[startz]

Hi again,

The build version the university uses is January 7 2010.

I also tried the same with Eviews 5 and I did not have any problems. So probably there is something with this version. I do not know if you have any other suggestion.

Ask whoever's in charge to run the free update. There have been numerous bug fixes and some free new features added.

[stasibab]

So should I assume that the problem would be with this specific version?

[startz]

Given that Gareth didn't have a problem, it's a reasonable bet that you've hit a bug that's already been fixed.

[Nau2306]

Hi

I am using eviews6. Can someone please tell me how to produce n-step ahead static garch forecasts? Can we make use of the above prgram suggested by Vaal1?

Thanks

[ysyzdy]

Dear users,

I m recently doing my dissertation and faced with problem in estimation basic rolling GARCh (1,1) process. I have 3650 observation and need to forecast 1 day ahead volatility in rolling form. I will highly appreciate if advanced users provide me assistance in that issue.. I looked guideline but there was no information. In addition I have 6 day remaining to finish my dissertation. So I need your help immediately. Please share any knowledge that you have in that issue

Jan

[CharlieEVIEWS]

There are examples on this forum which should get you going adequately:

http://forums.eviews.com/viewtopic.php?f=5&t=710