A Comparison of Models for Forecasting the Baltic Freight Index: Box-Jenkins Revisited

Abstract

The Baltic Freight Index (BFI) is a widely recognised barometer of dry bulk freight rates. As such, its composition is monitored continuously. In 1993, all handy size trades were expunged from the BFI. This paper tests whether the change in the composition of the BFI has altered its underlying behaviour. This is achieved by applying a Box-Jenkins methodology to a BFI database covering a period following this pivotal change and comparing the properties of the resulting ARIMA model to those of a model previously estimated by applying the same methodology to data from an earlier period. On the basis of a range of criteria, the two models prove to be remarkably similar and the paper concludes that the behaviour of the BFI has not been radically altered even following this radical revision.