Generalised Impulse Response Functions and Variance Decomp

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

jamest
Posts: 6
Joined: Wed Jul 13, 2011 9:14 am

Generalised Impulse Response Functions and Variance Decomp

Postby jamest » Mon Oct 17, 2011 7:46 am

Page II:470 of the user guide: "Note that since non-orthoganal factorization will yield decompositions that do not satisfy an adding up property, your choice of factorization is limited to the Cholesky orthogonal factorizations."

I don't really follow this.

If we are willing put up with the row sum of the elements of the variance decomposition matrix not summing to unity, then can we not use something other than Cholesky?

Basically I want to use the generalised impulses of Pesaran Shin to calculate the variance decomposition. Is this possible?

Thanks

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13307
Joined: Tue Sep 16, 2008 5:38 pm

Re: Generalised Impulse Response Functions and Variance Deco

Postby EViews Gareth » Mon Oct 17, 2011 7:53 am

No.
Follow us on Twitter @IHSEViews

jamest
Posts: 6
Joined: Wed Jul 13, 2011 9:14 am

Re: Generalised Impulse Response Functions and Variance Deco

Postby jamest » Mon Oct 17, 2011 8:20 am

I am none the wiser.

I am trying to do something based on this (pages 3 and 4). Is their expression (1) not a calculation of a variance decomposition based on the generalised impulses of Pesaran Shin?

http://www.ssc.upenn.edu/~fdiebold/papers/paper99/DirectionofSpillovers_Mar10.pdf

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13307
Joined: Tue Sep 16, 2008 5:38 pm

Re: Generalised Impulse Response Functions and Variance Deco

Postby EViews Gareth » Mon Oct 17, 2011 8:28 am

It is not built into EViews.
Follow us on Twitter @IHSEViews

jamest
Posts: 6
Joined: Wed Jul 13, 2011 9:14 am

Re: Generalised Impulse Response Functions and Variance Deco

Postby jamest » Mon Oct 17, 2011 8:55 am

And it's not possible to recover the necessary objects to construct it with a programme either?

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13307
Joined: Tue Sep 16, 2008 5:38 pm

Re: Generalised Impulse Response Functions and Variance Deco

Postby EViews Gareth » Mon Oct 17, 2011 9:04 am

I don't know enough about the specifics to judge, but I'd guess not.
Follow us on Twitter @IHSEViews

ppp007
Posts: 3
Joined: Sun Nov 20, 2011 5:39 am

Re: Generalised Impulse Response Functions and Variance Deco

Postby ppp007 » Sun Nov 20, 2011 5:41 am

Hi jamest,

I`m also trying to replicate Diebold paper, for a new set of time series. Did you find another way to run this?

Tks,

Paulo

jamest
Posts: 6
Joined: Wed Jul 13, 2011 9:14 am

Re: Generalised Impulse Response Functions and Variance Deco

Postby jamest » Fri Dec 16, 2011 6:22 am

Hi,
Only just seen this. I didn't make any further attempts but I may return to it in the New Year. How have you got on?
James

ppp007
Posts: 3
Joined: Sun Nov 20, 2011 5:39 am

Re: Generalised Impulse Response Functions and Variance Deco

Postby ppp007 » Wed Jan 04, 2012 6:49 pm

I will try to run N! ordering or, maybe, try running this on RATS.
I should check before which one is the less time demand.

nicktz
Posts: 18
Joined: Thu Mar 28, 2013 1:41 am

Re: Generalised Impulse Response Functions and Variance Deco

Postby nicktz » Fri May 24, 2013 3:12 am

Hi guys,

any of you any wiser on the technique by now? Help would be much appreciated!! N


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 29 guests