Multinomial backtesting of distortion risk measures
提出一种针对一般扭曲风险度量的回溯测试新方法，通过风险水平的分层与随机化来扩展可回溯测试的风险度量范围，并用数值案例验证其性能。
We extend the scope of risk measures for which backtesting methods are available by proposing a new approach for general distortion risk measures. The method relies on a stratification and randomization of risk levels. We illustrate the performance of our backtest in numerical case studies .