基于ARFIMA-HYGARCH-M-VaR模型的亚洲汇率市场均值和波动过程的双长期记忆性测度研究

Research on the Mean and Volatility of the Process of Double Long-term Memory Measure of Asian Currency Markets on ARFIMA-HYGARCH-M-VaR Model

作为金融传导机制的一个重要成分,汇率在金融危机的传播中发挥着重要作用.因此本文以亚洲汇率市场的汇率作为研究样本,通过引入skt分布来刻画残差的分布,构建了ARFIMA-HYGARCH-M-VaR模型来测度汇率风险值,并与skt分布下的GARCH及FIGARCH模型的VaR进行失败率回测检验与动态分位数测试.研究结果表明:在不同显著性水平下,skt分布下的各种模型基本都有较好的风险测度能力,且ARFIMA-HYGARCH-M模型的VaR风险测度更加精确与稳定.本研究为我国及亚洲其他国家汇率市场的风险测度与风险管理提供了一定的理论借鉴和方法基础.

This paper took the Asian exchange market exchange rates as the study sample. Using the skt-distribution to characterize the distribution of residuals, the ARFIMA-HYGARCH-M-VaR model was built to measure exchange rate risk value. The back test of failure rate and the dynamic quantile test have been taken by the GARCH of the skt-distribution and the VaR of the FIGARCH model. The results show that all kinds of models of the skt-distribution basically have a better risk measure capacity in the different significance level, and the risk measure capacity of the VaR of ARFIMA-HYGARCH-M model is more accurate and stable.