Financial Innovation

Table 9 Summary of the backtesting results for ablation studies

From: Forecasting VaR and ES by using deep quantile regression, GANs-based scenario generation, and heterogeneous market hypothesis

\(\tau\)	UC		CC		DQ		Mean (M-Loss)
\(\tau\)	0.01	0.05	0.01	0.05	0.01	0.05	0.01	0.05
Proposed Model	0	0	0	0	1	0	0.1379	0.3370
-(DA)	0	0	0	0	2	1	0.2629	0.5036
-(BO)	0	0	0	0	2	1	0.1862	0.4220
-(Mog)	0	0	0	0	3	1	0.1143	0.4774

-(DA), -(BO) and -(Mog) denote the removal of DA, BO and Mog from the proposed framework, respectively

Back to article page