Post by Sapphire Capital on Sept 8, 2008 10:51:36 GMT 4
Why VAR Models Fail and What Can Be Done
Elizabeth A. Sheedy
Macquarie Applied Finance Centre
August 29, 2008
Abstract:
During the sub-prime crisis of mid-2007, VaR models for market risk at many major financial institutions performed disappointingly. This performance is consistent with the use of VaR measures that fail to account for volatility clustering. This paper presents new backtesting evidence from equity markets to support the use of GARCH-based risk measures as a solution to this problem. The paper also examines many of the objections voiced by experts in relation to GARCH-based risk measures (such as the variability of such risk measures, regulatory issues, accessibility) and discusses possible solutions. Finally the paper examines some implementation issues related to GARCH-based VaR measures.
papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID1260655_code148688.pdf?abstractid=1260655&mirid=2
Elizabeth A. Sheedy
Macquarie Applied Finance Centre
August 29, 2008
Abstract:
During the sub-prime crisis of mid-2007, VaR models for market risk at many major financial institutions performed disappointingly. This performance is consistent with the use of VaR measures that fail to account for volatility clustering. This paper presents new backtesting evidence from equity markets to support the use of GARCH-based risk measures as a solution to this problem. The paper also examines many of the objections voiced by experts in relation to GARCH-based risk measures (such as the variability of such risk measures, regulatory issues, accessibility) and discusses possible solutions. Finally the paper examines some implementation issues related to GARCH-based VaR measures.
papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID1260655_code148688.pdf?abstractid=1260655&mirid=2