[R-SIG-Finance] number of observations - rugarch

[Patrick Burns]

Rather than answer the question, I'll talk past it.

I'm not convinced that it's worthwhile to worry about a garch model with 
less than 100 observations.  See:

http://www.portfolioprobe.com/2012/09/17/variability-of-garch-estimates/

and the blog post that is referenced in the update.

Good luck, Carolina, in any case.

Pat

On 15/10/2016 14:13, Carolina Magda Roma wrote:
> Dear all,
>
> First I want to say thanks for the beautiful rugarch package!
>
> I am trying to replicate some results and I would like to know if there is
> any adjustment that I can do to run the egarch model with external
> regressors using less than 100 observations. I tried to run my model but
> returns the message error below:
>
> library(rugarch)
> library(xts)
> data(sp500ret)
> spx<-xts(sp500ret, as.Date(rownames(sp500ret)))
> xreg<-xts(rnorm(nrow(spx)), index(spx))
> colnames(xreg)<-"xreg"
> xreg = lag(xreg,1)
> inputs<-na.omit(cbind(spx, xreg, join="left"))
>
> egarch.spec = ugarchspec(variance.model = list(model = 'eGARCH', garchOrder
> = c(1, 1)), mean.model = list(armaOrder = c(0, 0), include.mean = TRUE,
>
>     external.regressors= inputs[1:60,2]), distribution.model = 'sged')
> egarch.fit = ugarchfit(egarch.spec, inputs[1:60,1])
> egarch.fcst = ugarchforecast(egarch.fit, n.ahead=1)
>
> Error in .egarchfit(spec = spec, data = data, out.sample = out.sample,  :
>
> ugarchfit-->error: function requires at least 100 data
>  points to run
>
>
> Thanks a lot for any help!
>
> Carol Silva
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-SIG-Finance at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions should go.
>

-- 
Patrick Burns
patrick at burns-stat.com
http://www.burns-stat.com
http://www.portfolioprobe.com/blog
twitter: @burnsstat @portfolioprobe