Signup is easy. join now! | forgot password?

Bionic Turtle

Learn Finance with the pros. Better articles, resources and screencasts for easier learning.

Financial Risk Manger Exam Prep Register & Join our community (it's free!) Learn Free & Premium Access Articles with David Harper, CFA, FRM, CIPM More >
   
 
  • Home
  • Buy Now
  • Learn
  • About
  • Community
  • Premium
another Q re LiqVaR
 
You are here: Forum Home  >  Forums  >  Operational Risk  >  Thread
freemind
Posted: 12 November 2008 12:06 PM   [ Ignore ]  
Newbie
Rank
Total Posts:  3
Joined  2007-05-19

David,

In your question set on OpRisk B (question 5) you scale 1D LVaR to 10D LVaR using SQRT(10). Intuitively, I would only scale VaR (23,300 in the example) to 10D, and then add the same 1/2 spread as previously - as the spread is calculated on the full position of $1m and assumed constant at 0.3%.

so my solution would be: $1m * 0.01 * 2.33 * SQRT(10) + 0.5 * $1m * 0.003 = 75,181

Obviously this accounts only for the easy version of LVaR (ie w/o a stdev of spread given). Or do I miss something?

Thanks for enlightenment!

Michael

Profile
 
David Harper, CFA, FRM, CIPM
Posted: 12 November 2008 03:32 PM   [ Ignore ]   [ # 1 ]  
Administrator
RankRankRankRank
Total Posts:  1302
Joined  2006-09-24

Micheal,

I 100% agree with you (constant spread means don’t scale the spread with time).

I checked this against Dowd reference, he agree with you too: “liquidity adjustment...falls as holding period increases.”

My mistake. It is fixed...thanks for superb observation

David

Profile
 
   
 
 
‹‹ metallgesellschaft case      basel Credit ››

Powered By ExpressionEngine
Template Design based on Sonnenvogel.com | Customized by Lealea Design
Select a theme:

ExpressionEngine Discussion Forum - Version 2.0.0 (20070626)
Script Executed in 0.1568 seconds

RSS 2.0