Evaluating and Comparing Risk Model Performance
针对因子风险模型在预测波动率、优化投资组合和分解风险三方面的表现，提出了具体的检验方法，并指出常用评价指标的缺陷，对金融从业者选择风险模型有参考价值。
Factor risk models are used for three primary purposes: 1) predicting portfolio volatility, 2) portfolio optimization, and 3) decomposing risk and return into factor and idiosyncratic components. The authors propose specific tests or “horse races” to evaluate and compare the performance of competing risk models along these three basic dimensions. For evaluating the accuracy of volatility forecasts, they apply bias statistics and Q-statistics and stress the central importance of identifying the prediction horizon, which gives rise to the notion of the term structure of risk. For evaluating the efficacy of risk models for portfolio optimization, they argue strongly against the common practice of using realized information ratio as a measure. Instead, they advocate using the realized volatility of optimized portfolios, which they demonstrate is capable of correctly identifying the more efficient portfolio in a tiny fraction of the time. Finally, they argue that properly specified risk models must exhibit low correlations between systematic and idiosyncratic return contributions and demonstrate how the choice of regression weights has a significant role in decoupling factor from idiosyncratic returns.