A multicommodity model of futures prices: Using futures prices of one commodity to estimate the stochastic process of another

Authors

  • Gonzalo Cortazar,

    Corresponding author
    1. Departamento de Ingeniería Industrial y de Sistemas, Escuela de Ingeniería, Pontificia Universidad Católica de Chile in Santiago, Chile
    • Departamento de Ingeniería Industrial y de Sistemas, Escuela de Ingeniería, Pontificia Universidad Católica de Chile, Vicuña Mackenna 4860, Santiago, Chile
    Search for more papers by this author
  • Carlos Milla,

    1. FINlabUC, Laboratorio de Investigación Avanzada en Finanzas, Pontificia Universidad Católica de Chile in Santiago, Chile
    Search for more papers by this author
  • Felipe Severino

    1. FINlabUC, Laboratorio de Investigación Avanzada en Finanzas, Pontificia Universidad Católica de Chile in Santiago, Chile
    Search for more papers by this author

Abstract

This article proposes a multicommodity model of futures prices of more than one commodity that allows the use of long-maturity futures prices available for one commodity to estimate futures prices for the other. The model considers that commodity prices have common and commodity-specific factors. A procedure for choosing the number of both types of unobservable-Gaussian factors is presented. Also, it is shown how commodities with and without seasonality may be jointly modeled and how to estimate the model using Kalman filter. Results for the West Texas Intermediate–Brent and for the West Texas Intermediate–unleaded gasoline models presented show strong improvements over the traditional individual-commodity models, with much lower out-of-sample errors and better volatility estimates, even when using fewer factors. © 2008 Wiley Periodicals, Inc. Jrl Fut Mark 28:537–560, 2008

Ancillary