Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.802176
Title: Modelling volatility in energy markets
Author: Wang, Wenxue
Awarding Body: University of East Anglia
Current Institution: University of East Anglia
Date of Award: 2020
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Oil price uncertainty has a significant impact on economic growth and financial market performance, and understanding the drivers of oil price dynamics is vital for the global economy. It makes volatility modelling in crude oil pricing an essential topic for academics and practitioners. Geopolitical events in OPEC countries disrupting oil supply has long been the main driver of oil volatility, while the U.S. shale revolution has led to new dynamics on the supply side of the global oil market. The volatility transmission mechanism between crude oil and other assets in the investment market plays a crucial role in shaping international investments and economic policies. This thesis examines the channels of oil price dynamics and the volatility transmission mechanism between oil and other financial assets. Chapter 2 compares the performance of GARCH models, stochastic volatility models, and OVX implied volatility index regarding out-of-sample forecasting accuracy in oil futures prices. To do so, the dataset of West Texas Intermediate (WTI) oil active in the U.S. markets and the Brent Crude dominating the European market for the period 2004-2015 is adopted ,which includes the steep price drop in 2014. GJR-GARCH model suggests that leverage effect exists, while the stochastic volatility models are used to examine series dependence and heavy-tailed distribution in oil return series. The most important finding for this chapter is the detection of over - fitting in the GJR-GARCH model and stochastic volatility models. Chapter 3 examines whether the shale revolution has dampened the role of geopolitical risk in oil price volatility. Using the Structural Break Threshold Vector Autoregressive (SBT-VAR) framework proposed by Galvão (2006), this chapter identifies threshold effects and a structural break in April 2014. Furthermore, this study extends the framework of Galvão (2006) to a structural SBT-VAR system by allowing for conditional heteroskedasticity. Notably, impulse responses of oil price and (co)variance to the shock of geopolitical risk are compared over 170 periods, including before and after the shale oil revolution. We find that the impulse response functions of oil prices to a unitstructuralgeopoliticalriskshockhavebecomesmootherafterthebreakpoint in 2014 compared with those before the break. The main finding as we are expecting initially and intuitively is that the covariance response between geopolitical risk and oil price reduce with shale production shock compared to without. However, composing one extra unit of shale production shock makesthevolatilityresponseofoilpricestoageopoliticalriskshockreaching a higher level. Chapter 4 examines volatility spillovers and dynamic correlations between crude oil exchange-traded fund (ETF), various renewable energy ETFs, and the S&P 500 ETF by using multivariate GARCH-in-mean specifications. We find that the conditional volatility of the nuclear ETF has a significant positive effect on the oil ETF return, oppositely the volatility of the S&P 500 ETF negatively spillovers to the return of the oil ETF. The most important finding is that the long-term persistent volatility spillover from the S&P 500 ETF to renewable energy ETFs is significantly negative. Another result reveals that the dynamic correlations concurrently decrease before the financial crises (in 2008 and 2011 respectively) and then dramatically increase in the post-crisis period. Evidence shows that the dynamic correlation between oil ETF and S&P 500 ETF has always been positive since U.S. net imports of crude oil and petroleum products gradually decrease from 2005 onwards.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.802176  DOI: Not available
Share: