Abstract

In this paper, we propose an analytical approach to price a discretelysampled variance swap when the underlying asset is a commodity, with the realized variance defined in terms of squared percentage return of the underlying commodity prices. We assume that commodity price follows Schwartz (1997)'s one-factor model, which is adopted to describe the stochastic behavior of it. Furthermore, we demonstrate the validity of our closed-form solution in terms of its financial meaningfulness. Finally, a comparison between our solution and Monte Carlo simulations demonstrates the efficiency of our approach, which substantially reduces the computational burden of using Monte Carlo methods.