[R] function for prediting garch

[Kjetil Brinchmann Halvorsen]

oliver wee wrote:
> hello,
> 
> In my time series data, I was able to successfully fit
> its ARIMA model (Box-Jenkins) and its GARCH model and
> estimate their parameters. I was also able to forecast
> future values of the time series based on my fitted
> ARIMA model using the predict() function call. 
> 
> However, I'm not sure what is the correct function
> command to call in order to forecast  future values of
> my time series using both the fitted ARIMA model and
> the fitted GARCH model. Using predict() didn't give me
> the result I was looking for. And I can't find any
> documentation using help.search,

You should have given reproducible code!

In my understanding, (g)arch is applied to an
uncorrelated series without autocorrelastions,
as the residuals from a properly estimated ARIMA
model. So to get the predictions for the original
series, you need to
1) predict with the ARIMA model
2) estimate a garch model to the residuals
3) predict the residuals
4) modify the prediction from 1) with the prediction from 3)

Kjetil

> 
> I think what I am looking for is akin to the garchsim
> and garchpred commands in Mathlab.
> 
> Any help is appreciated. Thanks!
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>