CIRJE-F-618 "Pricing and Hedging of Long-term Futures and Forward Contracts by a Three-Factor Model"
Author Name Shiraya, Kenichiro and Akihiko Takahashi
Date April 2009
Full Paper  
Remarks Revised version of CIRJE-F-529 (2007); forthcoming in Quantitative Finance, page 1-16, Ahead of Print. Available online: 24 Mar 2010.
Abstract

This paper shows pricing and hedging efficiency of a three factor stochastic mean reversion Gaussian model of commodity prices using oil and copper futures and forward contracts. The model is estimated using NYMEX WTI (light sweet crude oil) and LME Copper futures prices and is shown to fit the data well. Furthermore, it shows how to hedge based on a three-factor model and confirms that using three different futures contracts to hedge long-term contract outperforms the traditional parallel hedge based on a single futures position by time series data and simulation. It also finds that the three factor model outperforms its two-factor version in replication of actual term structures and that stochastic mean reversion models outperform constant mean reversion models in Out of Sample hedges.