Connectedness between crude oil, coal, rare earth, new energy and technology markets: a GARCH-vine-copula-EVT analysis

ABSTRACT

In recent years, climate change has attracted great attention from governments and promoted the booming of the new energy market indirectly. However, this market will be influenced by traditional energy, rare earth and technology markets. Hence, it is necessary to incorporate these markets into an analytical framework simultaneously and analyse their relationships. Based on the GARCH-vine-copula-EVT model considering extreme risks, we investigate the connectedness between crude oil, coal, rare earth, new energy, and technology markets. The results show that the technology market is most closely associated with the new energy market; the rare earth market reacts as an intermediary market between the new energy market and fossil fuel markets. When taking the rare earth market as the conditional market, the connectedness between the new energy and the other four markets weakens and even becomes negative. Besides, we find that the COVID-19 epidemic has increased the connectedness between these target markets. Finally, the backtesting results of value at risk and expected shortfall show that the GARCH-vine-copula-EVT model considering extreme risks can depict the risk dependence structure between these target markets well. Our study has important reference significance for market participants, risk managers and investors.

KEYWORDS:

• Traditional energy
• new energy
• technology
• rare earth
• GARCH-vine-copula-EVT

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ According to Han, Liu, and Wang (2022) and Kumar et al. (2020), when determining the optimal copula functions between markets, the copula goodness of fit test and the AIC criterion are widely used. Manner (2007) suggests that the AIC criterion provides better performance than the goodness of fit test when selecting the copula function. And to verify the robustness of the copula functions selected by the AIC criterion, we also use the BIC criterion to select the optimal copula and the results are the same. Given that, we use the AIC criterion in this paper to select the optimal copula functions..