Re: [R] garch in tseries

From: Spencer Graves <spencer.graves_at_pdf.com>
Date: Thu 04 May 2006 - 02:19:11 EST

Dear Erin:

          The documentation for my currently installed version of garchFit{fSeries} indicates that the default include.mean = TRUE.

          How do you know "it didn't work"? With the default include.mean = TRUE, the following took just under a minute on my IBM T-30 notebook:

library(fSeries)
# Load Time Series:
data(sp500dge)
# Vector of Compounded Returns:
x <- diff(sp500dge[,1])
# Vector of Percentage Compounded Returns: X = 100*x
fit = garchFit(~arma(0,1), ~garch(1,1), series = X)

          Moreover, print(fit) included a parameter 'mu' whose names suggests that it is a mean that might not bre estimated if 'include.mean = FALSE'.

          However, when I tried it, it went into an apparent infinite loop, which I killed after ~15-20 minutes:

fit0 = garchFit(~arma(0,1), ~garch(1,1), series = X,

   include.mean=FALSE)

          I got a similar potentially infinite loop from the following:

fit2 = garchFit(~arma(0,1), ~garch(1,1), series = X+100)

          However, the following converged and estimated mu at 0.1000, as simulated:

fit.1 = garchFit(~arma(0,1), ~garch(1,1), series = X+0.1)

          Conclusion: The code seems to work as documented, sometimes. Unfortunately, it appears to have a disturbing affinity for an apparent infinite loop, and I don't have the time now to diagnose it further.

	  hope this helps.
	  spencer graves

Erin Hodgess wrote:

> Hello again!
>
> Is there a way to include a mean in the garch function in the
> library(tseries), please?
>
> I tried include.mean=T in the function statement but it didn't work
>
> thanks in advance!
>
> R Version 2.2.1 Windows
> Sincerely,
> Erin
> mailto: hodgess@gator.uhd.edu
>
> ______________________________________________
> R-help@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



R-help@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 Received on Thu May 04 03:39:14 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Thu 04 May 2006 - 04:09:58 EST.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.