Profitable Decision Rules in Mean Reverting Markets

Paolo Falbo
University of Brescia

Marco Frittelli
University of Milan

Silvana Stefani
University of Brescia
Stefani@master.cci.unibs.it

Abstract

We analyze the existence and characteristics of profitable decision rules in commodity markets. Evidence of mean reversion has been investigated and found in the recent literature on commodity price behavior. As shown in [Falbo, Frittelli, and Stefani1996], when the price process of a given commodity follows a mean reverting process, it is possible to determine trading strategies that performs better than others i.e. that offer expected returns higher than average with lower risk.

One of the purpose of this paper is to present a comparative analysis of these profitable strategies, in order detect optimal decision rules which maximize the expected return of the investment, for a given variance.

Using simulation techniques, we provide a description of these optimal strategies as a function of the relevant parameters of the model. Specifically, we show the relevance, for the determination of the optimal trading policy, of the ratio between the speed coefficient k of the mean reverting process and its volatility .

We estimate, for different commodities, the parameters of the mean reverting process and we then apply previous results to some specific commodities to show how to take, in practice, advantage of the presence of a high speed coefficient k.

Finally, we study how the mean reverting feature of commodities future price is related to the valuation of the on call contract provisions.

References

Falbo, Frittelli, and Stefani1996

Falbo, P., M. Frittelli, and S. Stefani, 1996. `Commodity futures markets and trading strategies opportunities', to appear in Modelling Techniques for Financial Markets and Bank Management, M. Bertocchi,  E. Cavalli, and S. Komlosi (Eds.), Physica Verlag, 48-64.