Forecasting the Real Price of Oil : A Cautionary Note

We study the out-of-sample predictability of the real price of crude oil using forecast combinations constructed from several individual predictors. We find that forecasts of themonthly average price of oil are more accurate than the no-change forecast at horizons ranging from 1 to 24 months with statistically significant MSFE reductions and directional accuracy. In contrast, we find no significant MSFE reductions or directional accuracy for the forecasts of the \textit{end-of-month} price of oil at almost all horizons. In application to hedging and investment decisions, we find that certainty equivalent values of hedging models and the Sharpe ratios of investment timing strategies that exploit predictability in monthly average prices are unduly high compared to similar strategies based on end-of-month price data. The implication of our results is that the statistical and economic significance of oil price forecastability is heavily influenced by the construction of the underlying oil price series and provide a cautionary note on which oil price series to use depending on the forecasting objective