Active portfolio management using robust optimization
研究了在概率分布和阈值双重不确定下，鲁棒期望损失和欧米伽比率优化模型，实证表明其在美国行业指数组合中显著跑赢基准和主动策略，并在高波动和系统性风险下表现更优。
Abstract We investigate robust models for Expected Shortfall (ES) and Omega Ratio (OR) optimization under joint uncertainty in both the probability distribution and the threshold. We apply this approach to actively manage portfolios comprising U.S. industry indices. Our empirical analysis shows that the robust ES and OR portfolios significantly outperform the benchmark index and active alternative strategies, even after adjusting for risk and transaction costs. Additionally, our findings demonstrate that the proposed robust optimization shifts allocations away from defensive sectors toward high-performing industries, capitalizing on upside-only momentum exposure and asset mispricing. Through simulation, we reveal that robust ES portfolios show pronounced advantages under high market volatility and cross-asset systematic risk variability, whereas robust OR portfolios benefit from low idiosyncratic volatility and notable asset mispricing. These findings underscore the effectiveness of robust ES and OR optimization in active portfolio management, highlighting their capacity to deliver strong performance and resilience under adverse market conditions.