Please use this identifier to cite or link to this item:
Härdle, Wolfgang
Stahl, Gerhard
Year of Publication: 
Series/Report no.: 
SFB 373 Discussion Paper No. 1999,105
VaR models are related to statistical forecast systems. Within that framework different forecast tasks including Value-at-Risk and shortfall are discussed and motivated. A backtesting method based on the shortfall is developed and applied to VaR forecasts of areal portfolio. The analysis shows that backtesting based on shortfall is very sensitive with respect to the underlying assumptions.
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
243.66 kB

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.