基于ARCH簇模型的即期汇率波动性研究

Analysis of Spot Exchange Rate Volatility Based on ARCH Clustering Models

建立了人民币/美元汇率波动的ARCH簇模型,分析汇率制度改革以来汇率波动特征,结果显示,汇率收益率具有尖峰厚望特征,收益率波动具有聚集性,建立了GARCH模型,得到汇率收益率是可预测的,可以利用过去的数据信息预测未来,在此基础上,建立TGARCH模型,表明汇率市场是不对称的,正面冲击和负面冲击对汇率的条件方差影响程度不同,又在TGARCH模型基础上建立TGARCH-M模型,表明市场汇率的收益与风险有关。

This paper used the ARCH（Autoregressive Conditional Heteroskedastic） models to analyze the variation characterstics of the CNY exchange rate against USD.The analytical results show that the CNY exchange rate is predictable and the past exchange rate can be used to predict the future.Then,based on the TGARCH model,we can see that the CNY exchange rate returns has an ARCH effect,and there are high peakmfat tail,volatility clustering and asymmetrics.The positive impact and negative impact are different on the exchange rate of the conditional variance.