[R-SIG-Finance] GARCH with variance targeting, stable distribution

Patrick Burns patrick at burns-stat.com
Mon Sep 10 11:18:40 CEST 2007

In answer to the second question, stable distributions
have infinite variance (except for the Gaussian), so garch
assuming a stable distribution is not particularly interesting.

Patrick Burns
patrick at burns-stat.com
+44 (0)20 8525 0696
(home of S Poetry and "A Guide for the Unwilling S User")

Jens Fricke wrote:

>Dear All,
>I am searching for a possibility to use the garch estimation with  
>variance targeting, i.e. first estimate the parameter a_0 in a garch  
>model (h_i=a_0+a_1*r_i-1^2+b_1*h_i-1) based on the unconditional  
>variance sigma^2: a_0=sigma^2*(1-a_1-b_1). Then, in a second step,  
>optimize the other parameters a_1, b_1. Is there a way to use the  
>"garch" routines or to incorporate that into the current functions?
>In addition, are there possibilities to estimate GARCH models with  
>assumed stable distribution?
>Thank you very much for your help.
>Best regards,
>R-SIG-Finance at stat.math.ethz.ch mailing list
>-- Subscriber-posting only. 
>-- If you want to post, subscribe first.

More information about the R-SIG-Finance mailing list