ALGORITHMS FOR OPTIMIZATION OF VALUE­ AT-RISK*

This paper suggests two new heuristic algorithms for optimization of Value-at-Risk (VaR). By definition, VaR is an estimate of the maximum portfolio loss during a standardized period with some confidence level. The optimization algo- rithms are based on the minimization of the closely related risk measure Conditional Value-at-Risk (CVaR). For continuous distributions, CVaR is the expected loss exceeding VaR, and is also known as Mean Excess Loss or Expected Shortfall. For discrete distributions, CVaR is the weighted average of VaR and losses exceeding VaR. CVaR is an upper bound for VaR, therefore, minimization of CVaR also reduces VaR. The algorithms are tested by minimizing the credit risk of a portfolio of emerging market bonds. Numerical experiments showed that the algorithms are efficient and can handle a large number of instruments and scenarios. However, calculations identified a deficiency of VaR risk measure, compared to CVaR. Minimization of VaR leads to an undesirable stretch of the tail of the distribution exceeding VaR. For portfolios with skewed distributions, such as credit risk, minimization of VaR may result in a significant increase of high losses exceeding VaR. For the credit risk problem studied in this paper, VaR minimization leads to about 16% increase of the average loss for the worst 1% scenarios (compared to the worst 1% scenarios in CVaR minimum solution). 1% includes 200 of 20000 scenarios, which were used for estimating credit risk in this case study.

[1]  Dirk Tasche,et al.  Risk contributions and performance measurement , 2000 .

[2]  D. Duffie,et al.  An Overview of Value at Risk , 1997 .

[3]  Claudia Klüppelberg,et al.  Optimal Portfolios with Bounded Capital at Risk , 2001 .

[4]  D. Tasche,et al.  On the coherence of expected shortfall , 2001, cond-mat/0104295.

[5]  Claudio Nordio,et al.  Expected Shortfall as a Tool for Financial Risk Management , 2001 .

[6]  John R. Birge,et al.  Introduction to Stochastic Programming , 1997 .

[7]  M. Pritsker.  Evaluating Value at Risk Methodologies: Accuracy versus Computational Time , 1996 .

[8]  S. Uryasev.  Probabilistic constrained optimization : methodology and applications , 2000 .

[9]  Didier Sornette,et al.  Have Your Cake and Eat It, Too: Increasing Returns While Lowering Large Risks! , 1999 .

[10]  Philippe Artzner,et al.  Coherent Measures of Risk , 1999 .

[11]  R. Rockafellar,et al.  Optimization of conditional value-at risk , 2000 .

[12]  Helmut Mausser,et al.  Applying Scenario Optimization to Portfolio Credit Risk , 2001 .

[13]  Helmut Mausser,et al.  Beyond VaR: from measuring risk to managing risk , 1999, Proceedings of the IEEE/IAFE 1999 Conference on Computational Intelligence for Financial Engineering (CIFEr) (IEEE Cat. No.99TH8408).

[14]  Helmut Mausser,et al.  Credit risk optimization with Conditional Value-at-Risk criterion , 2001, Math. Program..

[15]  Toshinao Yoshiba,et al.  On the Validity of Value-at-Risk: Comparative Analyses with Expected Shortfall , 2002 .

[16]  Peter Kall,et al.  Stochastic Programming , 1995 .

[17]  Amy v. Puelz,et al.  Value-at-Risk Based Portfolio Optimization , 2001 .

[18]  Yuri Ermoliev,et al.  Numerical techniques for stochastic optimization , 1988 .

[19]  J Figueira,et al.  Stochastic Programming , 1998, J. Oper. Res. Soc..

[20]  Andrey I. Kibzun,et al.  Stochastic Programming Problems with Probability and Quantile Functions , 1996 .

[21]  T. C. Wilson,et al.  Portfolio Credit Risk , 1998 .

[22]  A. Lucas,et al.  Extreme Returns, Downside Risk, and Optimal Asset Allocation , 1998 .

[23]  Dan Rosen,et al.  Credit Risk of an International Bond Portfolio: A Case Study , 2000 .

[24]  W. Ziemba,et al.  Worldwide asset and liability modeling , 1998 .

[25]  Olivier Scaillet,et al.  Sensitivity Analysis of Values at Risk , 2000 .

[26]  Helmut Mausser,et al.  Efficient Risk/Return Frontiers for Credit Risk , 2000 .

[27]  G. Pflug.  Some Remarks on the Value-at-Risk and the Conditional Value-at-Risk , 2000 .

[28]  M. O'connell.  Research Report 2001 , 2001 .