Market Risk Evaluation on Single Futures Contract: SV-CVaR Model and Its Application on Cu00 Data

A new stochastic volatility (SV) method to estimate the conditional value at risk (CVaR) is put forward. Firstly, it makes use of SV model to forecast the volatility of return. Secondly, the Markov chain Monte Carlo (MCMC) simulation and Gibbs sampling have been used to estimate the parameters in the SV model. Thirdly, in this model, CVaR calculation is immediate. In this way, the SV-CVaR model overcomes the drawbacks of the generalized autoregressive conditional heteroscedasticity value at risk (GARCH-VaR) model. Empirical study suggests that this model is better than GARCH-VaR model in this field.