Storability on Modeling Commodity Futures Prices

Econometric models of commodity prices have been estimated for more than 80 years, but both structural and time series models require ad hoc assumptions to capture all the features of commodity price series. Commodities can be broadly divided into two categories: storable and non-storable. The purpose of this study is to investigate the effects of storability on commodity futures pricing, especially whether meats can be reasonably approximated by storable commodity term structure models. From the empirical analysis of seven commodity futures prices, the two-factor Schwartz model is found to perform well for less storable commodities.