Centre for Global Finance Working Paper Series (ISSN 2041-1596) Paper Number: 02/10 Title: Measuring Shipping Tanker Freight Risk Author: Wessam Abouarghoub Centre for Global Finance Bristol Business School University of the West of England Coldharbour Lane Bristol BS16 1QY Telephone: 0117 32 83906 Email:
[email protected] Website: http://www.uwe.ac.uk/bbs/research/cgf/ MEASURING SHIPPING TANKER FREIGHT RISK WESSAM ABOUARGHOUB1 Bristol Business School, University of the West of England Coldharbour Lane, Bristol, BS16 1QY Tel: 01173281764 Mob: 07834069880
[email protected] 1 Lecturer & PhD candidate, Department of Economics, University of the West of England,
[email protected] 1 MEASURING SHIPPING TANKER FREIGHT RISK Abstract This is an attempt to measure level of risk exposure in tanker shipping freight markets by adapting a Value at Risk approach. These markets operate under conditions of perfect competition, and are extremely volatile, with clear presence of high volatility, seasonality and clusters in returns. They also exhibit leverage effects, and feature non-zero and high levels of skewness and kurtosis respectively. This study attempts to measure extreme conditional volatility through implementing different models that accommodate autoregression in conditional volatility of returns, leverage effects and fat-tails via parametric and semi-parametric specifications. In addition, Extreme Value Theory (EVT) is adopted to explicitly capture huge losses exhibited in the left-hand side of the distribution of returns, in comparison to numerous non-parametric approaches, Nonlinear-GARCH-t(d) and EVT-based models perform well in forecasting out-of-sample VaR. In addition, statistical tests show that the proposed models provide appropriate interval coverage in both unconditional and, more importantly, conditional contexts.