Value at Risk (VaR) has emerged as an useful tool to risk management. A relevant driving force has been the di usion as a benchmark of JP Mor- gan RiskMetrics methodology and the subsequent BIS adoption of VaR for all trading portfolios of nancial institutions. In this paper we propose rst an e cient and easy to implement algorithm to perform VaR estimation on the basis of orthogonal components, both in the equally weighted historical and in the exponentially weighted moving average method; secondly, incremental and marginal VaR expressions are derived in order to obtain quickly information about the change in portfolio risk when corrections in the composition are applied. Numerical results regarding time e ciency are presented and their implications are discussed.
Bramante, R., Decorrelation techniques in Value at Riskestimation, Contributed paper, in Proceedings of the Nineteenth International Conference - Forecasting Financial Markets, (marseille, 23-25 May 2012), FFM Conference, Marseille 2012: 1-7 [http://hdl.handle.net/10807/29392]
Decorrelation techniques in Value at Risk estimation
Bramante, Riccardo
2012
Abstract
Value at Risk (VaR) has emerged as an useful tool to risk management. A relevant driving force has been the di usion as a benchmark of JP Mor- gan RiskMetrics methodology and the subsequent BIS adoption of VaR for all trading portfolios of nancial institutions. In this paper we propose rst an e cient and easy to implement algorithm to perform VaR estimation on the basis of orthogonal components, both in the equally weighted historical and in the exponentially weighted moving average method; secondly, incremental and marginal VaR expressions are derived in order to obtain quickly information about the change in portfolio risk when corrections in the composition are applied. Numerical results regarding time e ciency are presented and their implications are discussed.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
