A Study On Exchange Rate Volatility Series Based On Wavelet Analysis

Reform of the RMB exchange rate system is a fundamental change for the RMB exchange rate mechanism, such changes helps maintain the RMB exchange rate at a more reasonable level. However, the RMB exchange rate determined by market will also increase the exchange rate risks. In order to guard against exchange rate risks effectively, the inherent nature of the exchange rate volatility series must be found out. The main purpose of this paper is to study the long-term memory of the exchange rate volatility series and the correlation between two different exchange rate volatility series.Firstly, a long memory analysis method based on wavelet variance for exchange rate volatility series is proposed. The exchange rate volatility series is analyzed with this method and the precise value of long memory parameters are obtained. The results show that various exchange rate volatility series has long memory property, and the greater the value of long memory parameters, the stronger effects of the historical information will the exchange rate volatility series suffer. Then, probability is used to measure two different exchange rate volatility series. As a result, the whole correlation is proved to be existed. A correlation analysis method based on wavelet covariance for different exchange rate volatility series was proposed, and the different exchange rate volatility series was analyzed with this method. The results show that the correlation between two different exchange rate volatility series in large-scale is stronger than in small-scale.