Benchmark Pricing of Weather Derivatives∗

This paper proposes an integrated approach to price weather derivatives based on the existence of an optimal benchmark portfolio for discrete time modelling. This portfolio, known as the growth optimal portfolio, when used as the numeraire ensures all benchmarked price processes are supermartingales. No further measure transformation is needed for the pricing of derivatives in a fair market, in the presence of completeness or incompleteness. An application of this approach in the pricing of weather derivatives is given. Since weather derivatives are not traded in a complete market, arbitragefree models cannot be used to determine fair prices, and therefore the use of actuarial pricing methods is required. A discrete time model is used to approximate historical values obtaining Gaussian distributed residuals. From this, fair weather contract prices are derived. JEL Classification: C15, G10, G13