[R-SIG-Finance] Incremental and Component VaR

Brian G. Peterson brian at braverock.com
Wed Dec 6 18:16:54 CET 2006

On Wednesday 06 December 2006 10:03, Joe W. Byers wrote:
> Brian,
> After reading your documented links.  The two source you are using
> confuse the definitions of IVAR and Marginal VAR as describe in the
> risk metrics documents.  IVAR in your first source is the riskmetrics
> definition. Marginal VAR in the second (sungard) is IVAR in the
> riskmetrics definitions and IVAR in this source in Marginal VAR in the
> riskmetrics definition.
> This means you are not calculating the IVAR correctly.
> Confusing isn't.

Prof. Byers,

Thank you for your comments on this matter.  I had previously noticed the 
definitional confusion on this, and thought (apparently erroneously) that 
I had sorted it out.  I'll re-review the references at my disposal and 
try again to unravel it.

I'm most interested in using a modified Cornish-Fisher VaR in these 
calculations, as I find that modified VaR is significantly more useful 
for assets that are not normally distributed than Riskmetrics style 
variance/covariance delta-normal VaR, but their descriptions are complete 
and coherent.  A number of authors have suggested that you can use 
modified VaR for Incremental/Component/Marginal VaR, so hopefully I'll be 
able to make it work with either traditional or modified VaR.

Thanks again for the review and pointers.  I'll post updates to this 
thread as I have them worked out.


   - Brian

> Brian G. Peterson wrote:
>> Ref:
>> Incremental and Component VaR is described many places in the
>> literature, including:
>> Beyond the VaR Horizon, Gaussel, et. al.
>> http://www.mathfin.com/nicolas/Q37.pdf
>> Incremental, Marginal, and Component VaR, 2004, Denton and Jayaraman
>> http://www.sungard.com/products_and_services/energy/variantsofvar1.pdf
>> extra_moments.R
>> http://braverock.com/brian/R/extra_moments.R

Return to Riskmetrics: the Evolution of a Standard
http://www.riskmetrics.com/pdf/rrmfinal.pdf (pp.69-72)

More information about the R-SIG-Finance mailing list