利用VAR模型来预测外汇储备各币种未来的收益率,利用DCC-GARCH模型结合条件风险价值CVaR动态地描述不同币种结构下外汇储备的风险状况。为了突出极端风险控制,通过建立均值一CVaR模型,研究不同目标收益率要求下的中国外汇储备币种的动态最优结构。结果表明,在不降低收益率的前提下,动态配置后的外汇储备CVaR风险显著降低。
We analyze the volatility of the exchange rate of various foreign exchange reserves based on VAR model and present dynamically the risk of the different currency structure of foreign exchange reserves by using DCC- GARCH model combining with the conditional value at risk(CVaR). Ultimately, according to Markowitz mean-vari- ance model, we propose the mean-CVaR model to study the dynamic optimal currency structure of China's foreign ex- change reserves under different target yields. The results show that dynamic allocation can reduce the CVaR risk of foreign exchange reserves significantly at the same given yield rate.