基于SV-POT模型的黄金市场的动态VaR预测

VaR Forecasting for Gold Market Based on SV-POT Model

针对黄金市场呈现的"尖峰厚尾"和波动持续性等特征,选用SV(stochastic volatility)模型来刻画。将SV模型与基于POT(peak over threshold)模型的极值理论相结合,建立SVPOT的组合模型,预测该金融市场的动态VaR(value at risk)。最后,与GARCH-POT模型相比得出:基于随机波动模型的SV-POT模型在一定程度上能更精确地预测动态VaR。

The SV （ Stochastic Volatility） model is used to depict the characteristics of the ＂fat tail＂ and volatility persistence in the gold market. Then combining the SV model with the extreme value theory based on POT （ Peak Over） model, the combination model of SV-POT is established to predict the dynamic VaR of the financial market. Finally, compared with the GARCH-POT model, the SV- POT model based on stochastic volatility model can predict VaR （Value at Risk） more accurately in a certain extent.