Backtesting Systemic Risk Forecasts Using Multi-Objective Elicitability
指出CoVaR、CoES和MES等系统性风险度量指标无法被直接验证，并提出多目标可引性概念，利用二维评分和字典序的Diebold-Mariano检验来解决回测问题，最后用DAX 30和S&P 500数据演示了交通灯方法。
Systemic risk measures such as CoVaR, CoES and MES are widely-used in\nfinance, macroeconomics and by regulatory bodies. Despite their importance, we\nshow that they fail to be elicitable and identifiable. This renders forecast\ncomparison and validation, commonly summarised as `backtesting', impossible.\nThe novel notion of \\emph{multi-objective elicitability} solves this problem.\nSpecifically, we propose Diebold--Mariano type tests utilising two-dimensional\nscores equipped with the lexicographic order. We illustrate the test decisions\nby an easy-to-apply traffic-light approach. We apply our traffic-light approach\nto DAX~30 and S\\&P~500 returns, and infer some recommendations for regulators.\n