This paper proposes a method of estimating Value at Risk (VaR) based on the assumption that the financial returns follow a switching regime ARCH model. We use the simple switching-regime model, the traditional GARCH(1,1) model and the switching-regime ARCH model to do some empirical analysis and to calculate the VaR values under different confidence levels for Shanghai and Shenzhen Stock Index. The calculated VaR values are com- pared. The results of back-testing and the Proportion of Failure test show the VaR values calculated by the switching-regime ARCH model are preferred to other methods.
|Number of pages||18|
|Journal||Journal of Industrial and Management Optimization|
|Publication status||Published - 2006|
- switching regime
- ARCH model
- fat-tailed distribution
- modelling science