Worst-Case Conditional Value-at-Risk with Application to Robust Portfolio Management

This paper considers the worst-case CVaR in situation where only partial information on the underlying probability distribution is given. It is shown that, like CVaR, worst-case CVaR remains a coherent risk measure. The minimization of worst-case CVaR under mixture distribution uncertainty, box uncertainty and ellipsoidal uncertainty are investigated. The application of worst-case CVaR to robust portfolio optimization is proposed, and the corresponding problems are cast as linear programs and second-order cone programs which can be efficiently solved. Market data simulation and Monte Carlo simulation examples are presented to illustrate the methods. Our approaches can be applied in many situations, including those outside of financial risk management.