Nonlinear Test and Forecasting of Petroleum Futures Prices Time Series

AbstractIn this study, an attempt is made to characterize and predict petroleum futures prices, using ideas gained from nonlinear dynamical theory. Data on the New York Mercantile Exchange (NYMEX) of crude oil and heating oil futures are used. The R/S analysis, the power spectrum and the largest Lyapunov exponent are in close agreement with each other, providing convincing evidence regarding the presence of chaotic behavior in the daily petroleum futures prices series. Nonlinear forecast modeling based on phase space reconstruction is applied to petroleum futures prices series. The results indicate the appropriateness of the nonlinear dynamical approach for characterizing and predicting the dynamics of petroleum futures prices