Multiple Timescale Analysis of Metal Prices Volatility Based on Empirical Mode Decomposition
The sustained run-up and strong metal prices volatility cause uncertainty to producers, consumers, stockholders and attract more and more attention recently. The metal prices, however, are non-linear and non-stationary financial time series. It is difficult for us to understand the underlying characteristics of metal prices volatility with traditional methods and necessary to find some more adaptive techniques. Therefore, this study applies the empirical mode decomposition (EMD) method to decompose and restructure the futures prices of three metals (Cu, Al, Ag) traded in London Metal Exchange into three components: short-term, medium-term and long-term. Then, the correlation and economic meanings analysis of the resulting components are performed from multiple timescale perspective. Empirical analysis indicates that, first, long-term and medium-term components are deterministic forces for metal prices; second, there exists a bidirectional causality relationship between two of the three metals prices for long-term or medium-term component, but only silver price has effect on copper price for short-term components; third, there are some evidences of a bidirectional volatility spillover between copper and aluminum in short-term timescale. This study extends the EMD method to the metal prices research field and provides an effective price forecasting foundation for further research.
metal prices empirical mode decomposition multiple timescale
Jianbo Huang Hui Cheng Yaoqi Guo
School of Business,Central South University,Changsha 410083,China
国内会议
长沙
英文
1-9
2012-10-11（万方平台首次上网日期，不代表论文的发表时间）