Improving portfolio risk profile with threshold accepting

The application of the Threshold Accepting (TA) algorithm in portfolio optimisation can reduce portfolio risk compared with a Trust-Region local search algorithm. In a benchmark comparison of several different objective functions combined with different optimisation routines, we show that the TA search algorithm applied to a Conditional Value at Risk (CVaR) objective function yields the lowest Basel III market risk capital requirements. Not only does the TA algorithm outmatch the Trust-Region algorithm in all risk and performance measures, but when combined with a CVaR or 1% VaR objective function, it also achieves the best portfolio risk profile.

[1]  Gerhard W. Dueck,et al.  Threshold accepting: a general purpose optimization algorithm appearing superior to simulated anneal , 1990 .

[2]  V. Bawa.  OPTIMAL, RULES FOR ORDERING UNCERTAIN PROSPECTS+ , 1975 .

[3]  Peter F. CHRISTOFFERSENti.  EVALUATING INTERVAL FORECASTS , 2016 .

[4]  A. Roy.  SAFETY-FIRST AND HOLDING OF ASSETS , 1952 .

[5]  Manfred Gilli,et al.  A Data-Driven Optimization Heuristic for Downside Risk Minimization , 2006 .

[6]  Richard H. Byrd,et al.  A Trust Region Algorithm for Nonlinearly Constrained Optimization , 1987 .

[7]  Peter Winker,et al.  New concepts and algorithms for portfolio choice , 1992 .

[8]  D. Tasche,et al.  Expected Shortfall: a natural coherent alternative to Value at Risk , 2001, cond-mat/0105191.

[9]  Peter Christoffersen,et al.  Elements of Financial Risk Management , 2003 .

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

[11]  Yannick Malevergne,et al.  Alternative risk measures for alternative investments , 2006 .

[12]  Dietmar Maringer,et al.  Portfolio management with heuristic optimization , 2005 .

[13]  Enrico Schumann,et al.  Numerical Methods and Optimization in Finance , 2011 .

[14]  Peter Winker.  Optimization Heuristics in Econometrics : Applications of Threshold Accepting , 2000 .

[15]  Phhilippe Jorion.  Value at Risk: The New Benchmark for Managing Financial Risk , 2000 .

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

[17]  Russell C. Eberhart,et al.  A new optimizer using particle swarm theory , 1995, MHS'95. Proceedings of the Sixth International Symposium on Micro Machine and Human Science.

[18]  A. Roy.  Safety first and the holding of assetts , 1952 .

[19]  T. Coleman,et al.  Minimizing CVaR and VaR for a portfolio of derivatives , 2006 .

[20]  SUPERVISORY FRAMEWORK FOR THE USE OF "BACKTESTING" IN CONJUNCTION WITH THE INTERNAL MODELS APPROACH TO MARKET RISK CAPITAL REQUIREMENTS , 1996 .

[21]  Luca Maria Gambardella,et al.  Ant Algorithms for Discrete Optimization , 1999, Artificial Life.

[22]  R. Cont.  Empirical properties of asset returns: stylized facts and statistical issues , 2001 .

[23]  A. Stuart,et al.  Portfolio Selection: Efficient Diversification of Investments , 1959 .

[24]  Manfred Gilli,et al.  A Data-Driven Optimization Heuristic for Downside Risk Minimization , 2006 .

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

[26]  E. Fama.  The Behavior of Stock-Market Prices , 1965 .