Do spot food commodity and oil prices predict futures prices?

摘要

Futures prices reflect the price that both the buyer and the seller agree will be the price of a commodity upon delivery. Therefore, these prices provide direct information about investor's expectations about the future price of the commodity of interest. This purpose of this research is twofold. First, following earlier investigations, an effort is made to understand the extent to which the spot energy price contains information content in the current period useful for predicting the forward-looking variable. The working hypothesis is that both own-commodity spot prices and spot energy prices are significant predictors of future commodity prices at alternative leads (lags). Second, the research investigates the predictive accuracy and biasedness of futures prices predictions from reverse regressions using in-sample criteria as well as from the performance of the models based upon ex post forecasts generated by alternative time series models. The results indicate that in some cases spot own-commodity prices and spot oil prices are useful for predicting prices of futures contracts although the lead-lag relationships vary considerably as between commodities and markets considered as well as with respect to temporal aggregation. Further, the evidence suggests that unless there is specific interest in the EGARCH parameter estimates, GARCH models tend to perform at least as well as without the added complexity of EGARCH.