Risk spillover measurement of carbon trading market considering susceptible factors: A network perspective

Abstract

An objective and robust network-based data-driven strategy is proposed to analyze risk spillovers in carbon markets. First, we characterize the causality network between the carbon market and potential associated markets using a data-driven fuzzy cognitive map approach. Second, network-based community detection is conducted to explore community structures that include carbon trading markets, and five market factors belonging to the same community as EU Allowances (EUA) are identified. Next, we conduct downside and upside-tail measurements of EUA risk spillover levels within the community based on estimates and fits of marginal and joint distributions for different market pairs. Finally, we point out that the market factor having the most significant upper-tail spillover effects on EUA is OILFUTURE, besides, EURUSD asset is found to be the best hedge for EUA futures among the detected market factors.