Measuring extreme risk dependence between the oil and gas markets
- PDF / 2,112,052 Bytes
- 18 Pages / 439.37 x 666.142 pts Page_size
- 94 Downloads / 227 Views
Measuring extreme risk dependence between the oil and gas markets Hachmi Ben Ameur1
· Zied Ftiti2 · Fredj Jawadi3 · Wael Louhichi4
Accepted: 9 September 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract This study aims to measure the risk dependence between the two most important energy markets, oil and gas, to analyze their risk spillovers. To this end, we apply different extreme risk measures (the value at risk, conditional value at risk, delta conditional value at risk, and copula) to high-frequency energy data to capture the intraday dynamic dependence between oil and gas prices (using, in particular, a 5-min intraday sample data from November 2014 to October 2017). Our analysis shows two interesting findings. First, while we highlight an extreme risk dependence between oil and gas markets, the risk spillover from the oil to the gas market is higher than that from the gas to the oil market. Second, the upward and downward risk spillovers exhibit asymmetry, as extreme negative shocks produce a stronger spillover effect than do extreme positive shocks. The exploration of these systemic risk forms provides significant insights for policymakers and investors in terms of risk management and portfolio diversification. Keywords Systemic risk · VaR · CoVar · Dynamic copula · Intraday data JEL Classification C22 · G1 · Q4
1 Introduction Gas and oil are considered two of the most important commodities, not only given their key roles for consumption and production functions, but also with regard to their liquidities and trading volumes. Further, over the last decades, they have increasingly captured the
B
Hachmi Ben Ameur [email protected]
1
INSEEC Grande Ecole, INSEEC U, Paris, France
2
OCRE-Laboratory, EDC Paris Business School, Paris, France
3
University of Lille, Lille, France
4
ESSCA School of Management, Paris, France
123
Annals of Operations Research
attention of investors, policymakers, and scholars in reason of their extreme excess volatilities. Accordingly, the investigation of the relationship between these two commodity markets is a central issue in the energy literature that might provide useful insights. The analysis of this relationship is useful for policymakers, investors, and hedgers in terms of portfolio diversification, market regulation, and price forecasting. Obviously, from an economic point of view, gas and oil might correlate, as both commodities appear to be substitutes or complementary assets. Empirically, the investigation of the gas-oil relationship is not a new issue and has been at the center of several related empirical studies (Bacon 1991; Karrenbrock 1991; Faff and Brailsford 1999; Sadorsky 2001; Boyer and Filion 2007; Nandha and Faff 2008; Brigida 2014; Lin and Li 2015; Batten et al. 2017; Ftiti et al. 2020).1 Overall, these studies tested the oil–gas relationship using either classical timeseries models (vector autoregression family models, co-integration, error correction model, and the Granger causality analysis), nonlinear time-series
Data Loading...
