s-news
[Top] [All Lists]

summary: innovations algorithm for time series prediction

To: s-news@lists.biostat.wustl.edu
Subject: summary: innovations algorithm for time series prediction
From: Vera Friederichs <vera@statlab.uni-heidelberg.de>
Date: Thu, 22 Nov 2001 14:50:12 +0100
References: <3BFA32A8.3D724180@statlab.uni-heidelberg.de> <Pine.GSO.4.10.10111200710430.5977-100000@phz-6>
Reply-to: vera@statlab.uni-heidelberg.de
Thank you to all who replied. For our purpose Matt Calder's suggestion
is the most useful.

Best regards,
Vera Friederichs

Matt Calder wrote:
> 
> Vera,
>         Do you have a recent copy of PEST that comes with the Brockwell and
> Davis book? This can import and export data from most Windows programs, and
> provides a friendly GUI to guide analysis. If you are looking for
> automatic/programmatic solutions, try writing the innovations algorithm in S, 
> it
> is pretty straightforward, and would be reasonalbly efficient.
> 
>         Matt
> 
> On Tue, 20 Nov 2001, Vera Friederichs wrote:
> 
> > Dear list,
> >
> > I am urgently searching an implementation of the classical innovations
> > algorithm (see, e.g. Brockwell and Davis, Time Series: Theory and
> > Methods) for time series prediction to be used with S-Plus (which
> > unfortunately doesn't have this algorithm
> > implemented). I guess the easiest would be if somebody knew about some
> > C or Fortran code which could be directly linked to S-Plus?
> >
> > Thanks,
> > Vera Friederichs
> > ---------------------------------------------------------------------
> > This message was distributed by s-news@lists.biostat.wustl.edu.  To
> > unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
> > the BODY of the message:  unsubscribe s-news
> >

<Prev in Thread] Current Thread [Next in Thread>