[18898] in s-news-athena
[S] [S+GARCH] Correlation Forecast
daemon@ATHENA.MIT.EDU (=?iso-8859-1?q?Manos=20Venardos?=)
Tue Aug 10 04:27:31 1999
Message-Id: <19990810081809.25361.rocketmail@web601.yahoomail.com>
Date: Tue, 10 Aug 1999 09:18:09 +0100 (BST)
From: =?iso-8859-1?q?Manos=20Venardos?= <evenos@yahoo.co.uk>
To: s-news@wubios.wustl.edu
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
I am trying to fit a multivariate dvec(1,1) or bekk (1,1) model for 3
variables using
mod<-mgarch(data~1,~dvec(1,1)) OR
mod<-mgarch(data~1,~bekk(1,1))
The problem is: when using the command
predict(mod,X)
to forecast my estimated model X periods ahead, the correlation
forecasts frequently exceed 1! Does anybody know why this happens? Note
that I have never encountered this problem when fitting bivariate dvec
or bekk models. It only happens when the number of variables is grater
than 2.
Manos Venardos
University of Oxford
Nuffield College
_____________________________________________________________
DO YOU YAHOO!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
