Research On Fractal Feature Of European Carbon Market

China is in the rapid development stage of the industrialization and urbanization, controling carbon dioxide emissions has become an urgent task following by a significant increase in carbon emissions. Emissions of carbon dioxide and other greenhouse gases promote carbon trading market to become another very important financial markets in the world. Many developing countries of the EU and other developed countries have established a carbon trading market since the signing of the "Kyoto Protocol". The EU ETS’s carbon trading system is relatively mature and the largest carbon market among them, it is in a dominant position in the global carbon market. Although carbon emission exchange markets have established in China, they are still in their infancy period, they have the small volume, there are not a national carbon emissions price mechanism. Understanding the EU ETS’s carbon price volatility characteristics and grasping carbon price volatility risk are an important reference for the future development of China’s carbon trading market.Through doing statistical description of EUA and CER in carbon market, thinked that the return series had a fat tail, did not obey normal distribution,and can not be analyzed by the efficient market hypothesis, fractal theory accorded with carbon market features.Fractal theory had many advantages in describing the characteristics of carbon price volatility, which had the detailed analysis of the complex system from both local and global aspects. Nonlinear characteristics of the carbon market were analyzed from single fractal and multi-fractal aspects. First, the single fractal characteristics of EU ETS was analyzed overally using V/S and DFA methods, and found that the second transaction phase of EUA and CER existed long-term memory, the third phase was the anti-state persistent presence, Hurst index was near0.5, the structure of EUA and CER time series is more complex and market volatility was relatively large, investment venture of EUA and CER were relatively high, calculating the average cycle of EUA and CER basing on the analysis of the V statistic, that was how long the current information generated impact on the future, the cycle period of EUA two phase were355days and255days, the cycle of CER were360days and158days. According to the,the cycle time, investors can hold the carbon market to do a certain degree, reduced market risk.Secondly, the multifractal spectrum of EUA and CER return time series were analyzed using a sliding window MF-DFA and MF-DMA methods, which found both existed multi-fractal characteristics, but CER was more sifnificant, that is, CER had more larger investment risk than EUA. In addition, producing multifractal reasons of the original sequence was analyzed after rearraangement and randomized treatment, thank that long-term memory and thick tail distribution together lead to the existence of multiple fractal.Finally, considing the long-run equilibrium relationship between EUA and CER, using MF-DXA method to analyze cross-correlation between EUA and CER, the empirical results showed that multi-fractal characteristics existed between them, and the second transactions phase was more evident, price volatility was more intenser than the third phase,indicating that the carbon trading market became more mature.