[R-SIG-Finance] [R-sig-finance] practicability of variance decomposition

[Eric Zivot]

The usefulness of variance decompositions is described in several time series books - Lutkepohl, enders, etc. Favero's applied macroeconometrics book is also very good in this regard. The idea is to see what are the sources of forecast errors. If you are predicting real GDP from a SVAR with 5 variables and 5 identified orthogonal "shocks" then the FEVD tells you which variable is contributing most to the forecast error. Lutkephoh also shows very clearly that the FEVD gives the same information as the impulse response functions but it a different way. FEVD are not particularly useful outside of an identified SVAR because the decomposition of variance only has interpretable meaning when the shocks are uncorrelated.

****************************************************************
*  Eric Zivot                  			               *
*  Professor and Gary Waterman Distinguished Scholar           *
*  Department of Economics                                     *
*  Adjunct Professor of Finance                                *
*  Adjunct Professor of Statistics
*  Box 353330                  email:  ezivot at u.washington.edu *
*  University of Washington    phone:  206-543-6715            *
*  Seattle, WA 98195-3330                                      *                                                           *
*  www:  http://faculty.washington.edu/ezivot                  *
****************************************************************

On Thu, 16 Jul 2009, RON70 wrote:

>
> Hi all, in multi-variate TS literature, in terms of VAR, VECM etc, I see lot
> of stuffs on Variance decomposition, however could not figure out what is
> the practical use of that? What practical question it answers? Any idea?
>
> Regards,
> --
> View this message in context: http://www.nabble.com/practicability-of-variance-decomposition-tp24520555p24520555.html
> Sent from the Rmetrics mailing list archive at Nabble.com.
>
> _______________________________________________
> R-SIG-Finance at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only.
> -- If you want to post, subscribe first.
>