Extreme Risk Measurement of Carbon Market Considering Multifractal Characteristics

Influenced by the global economy, politics, energy and other factors, the price of carbon market fluctuates sharply. It is of great practical significance to explore a suitable measurement method of extreme risk of carbon market. Considering that the return series of carbon market has the characteristics of leptokurtosis, fat tail, skewness and multifractal, and there maybe many extreme risk values in the carbon market, this paper introduces the Skewed- t distribution which can describe the characteristics of leptokurtosis, fat tail and skewness of return series into MSM model which can describe multifractal characteristic of return series to model volatility of carbon market. On the basis, based on the extreme value theory, this paper constructs Skewed- t -MSM-EVT model to measure extreme risk of carbon market. This paper chooses EUA market as the object to study extreme risk of carbon market, and draws the following conclusions: Skewed- t -MSM-EVT model has significantly higher prediction accuracy for carbon market’s VaR than MSM-EVT models under other distributions (including normal distribution, t distribution, GED distribution); Skewed- t -MSM-EVT model is superior to traditional Skewed- t -FIGARCH-EVT and Skewed- t -GARCH-EVT models in predicting carbon market’s VaR. This research has important practical significance for accurately grasping the risk of carbon market and promoting energy conservation and emission reduction.