Standard and comparative e-backtests for general risk measures

Abstract: Backtesting risk measures is a unique and important problem for financial regulators to evaluate risk forecasts reported by financial institutions. As a natural extension to standard (or traditional) backtests, comparative backtests are introduced to evaluate different forecasts against regulatory standard models. Based on recently developed concepts of e-values and e-processes, we focus on how standard and comparative backtests can be manipulated in financial regulation by constructing e-processes. We design a model-free (non-parametric) method for standard backtests of identifiable risk measures and comparative backtests of elicitable risk measures. Our e-backtests are applicable to a wide range of common risk measures including the mean, the variance, the Value-at-Risk, the Expected Shortfall, and the expectile. Our results are illustrated by ample simulation studies and real data analysis.