Volatility forecasting across tanker freight rates: the role of oil price shocks

Abstract

This paper examines whether the inclusion of oil price shocks of different origin as exogenous variables in a wide set of GARCH-X models improves the accuracy of their volatility forecasts for spot and 1-year time-charter tanker freight rates. Kilian's (2009) oil price shocks of different origin enter GARCH-X models which, among other stylized facts of the tanker freight rates examined, take into account the presence of asymmetric and long-memory effects. The results reveal that the inclusion of aggregate oil demand and oil-specific (precautionary) demand shocks improves significantly the accuracy of the volatility forecasts drawn.