Modeling and Forecasting the Volatility of Oil Futures Using the ARCH Family Models

Abstract

This study attempts to model and forecast the volatility of light, sweet, crude oil futures trading at the NYMEX during 1998–2009, using various models from the ARCH family. The results reveal that the GJR-GARCH (1,2) model is best suited to forecast purposes. The fitted models also suggest the presence of asymmetric effects in the data. The study also reveals that trading volume and open interest do not reduce the persistence of volatility for these oil futures.