Regional and copula estimation effects on EU and US energy equity portfolios

ABSTRACT

The aim of this paper is to investigate the regional interdependence structure of energy equities in the US and in the EU. Based on weekly stock prices of 28 big energy firms in the two regions from 2008 to 2019, we compare the efficiency of using bivariate or multivariate copulas to describe the dependence structure of energy equities. Furthermore, we investigate the impact of the choice between these two methods on the performance of energy equity portfolios. Our empirical results show that multivariate copulas, such as C-Vine, allow to better describe the dependence structure of energy equities. We also find that there is a stronger and more complex dependence structure among EU energy equities than among US energy equities. Our scenario analysis also shows that the dependence structure is stronger during the GFC while being weaker during the ESDC. More importantly, the correlation matrix obtained from the multivariate copula method allows to obtain optimal mean-CVaR portfolios with a higher performance than that from the bivariate copula method. More importantly, optimal portfolios constituted with multivariate copulas allow to reduce the portfolio’s sensitivity to oil prices.

KEYWORDS:

• Portfolio optimization
• energy equities
• copulas
• oil

JEL CLASSIFICATION:

• C16
• C58
• F36
• G11
• Q40

Highlights

• Multivariate copulas describe better the dependence structure of energy equities.

• The dependence structure is stronger among EU energy equities than among US energy equities.

• Multivariate copulas allow to construct optimal portfolios with higher performance.

• Multivariate copulas allow to reduce the portfolio’s sensitivity to oil prices.

Acknowledgments

We would like to thank the Editor of Applied Economics, Mark P. Taylor and an anonymous reviewer for their valuable comments that allow us to improve our manuscript significantly. Any shortcoming or error remains the authors’ responsibility. This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2017S1A5B8057488).

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2017S1A5B8057488).

² An analysis about the behaviour of oil prices can be found in Reboredo, Quintela, and Otero (2017a).

³ This scenario analysis is inspired by the approach of Giuzio et al. (2019).

Additional information

Funding

This work was supported by the National Research Foundation of Korea [NRF-2017S1A5B8057488].