[R-SIG-Finance] Seasonal GARCH

[Gabor Grothendieck]

On Sun, Apr 6, 2008 at 9:49 PM,  <markleeds at verizon.net> wrote:
> >From: Spencer Graves <spencer.graves at pdf.com>
> >Date: 2008/04/06 Sun PM 08:38:58 CDT
> >To: ihernan at stat.berkeley.edu
> >Cc: r-sig-finance at stat.math.ethz.ch
> >Subject: Re: [R-SIG-Finance] Seasonal GARCH
>
> I've never used garchFit but one other possibility
> is to use the fact that an ARCH model of the volatility
> is the same as an AR whatever on sigma squared
> , with some subtle, slight differences that I can't
> recall. So, using that relation, you may be
> able to use an AR ( so one of the arima functions )
> but be careful because it doesn't imply EXACTLY
> the same model but it's close.
>
> I forget which text talks about the close but no cigar relation but my guess is that it's in Hamilton's or Zivot's text.
>

Its mentioned in Lutkepohl's book, New Intro to
Multiple Time Series Analysis.