Measuring the Market Risk of Freight Rates: A Forecast Combination Approach

Abstract

This paper aims at contributing to the literature in three ways: First, we re-evaluate the performance of popular Value-at-Risk (VaR) estimation methods on freight rates amid the adverse economic consequences of the recent financial and sovereign debt crisis. Secondly we provide a detailed and extensive backtesting methodology in order to identify possible weaknesses associated with the standard backtesting criteria. A newly proposed method is employed in order to evaluate the performance of each method. Last, we propose a combination forecast approach for estimating VaR. Our findings suggest that both the parametric and simulation methods produce accurate estimates of daily VaR. More importantly, our combination methods produce more accurate estimates at both the 1% and the 5% significance level and for all the sectors under scrutiny, while in some cases they may be viewed as conservative since they tend to overestimate nominal VaR.

Keywords: Backtesting, Combination Forecasts, Volatility Forecasts, Freight Rates, Performance Evaluation, Value-at-Risk

JEL Classification: G10, G13