A consistent two-factor model for pricing temperature derivatives

- We provide a consistent two-factor model for pricing temperature derivatives that incorporates the forward looking information available in the market.
- The model specifies the dynamics of the complete meteorological forecast curve reasonably well.
- We analyze the model's performance in an application using real temperature forecast data and temperature futures derivative prices.
- The model confirms that parts of the irregularity of the market price of risk is due to information misspecification.
- Similar to temperature derivatives, this approach can be used for pricing other non-tradable assets.

In the past decade, the Chicago Mercantile Exchange began to trade weather derivatives to hedge weather risk. The pricing of weather derivatives is challenging since the underlying is not tradable and thus classical arbitrage approaches have to be used with caution. In typical pricing approaches all information available to the market is assumed to be incorporated in the underlying and thus forward-looking information about non-tradable assets such as meteorological forecasts is often ignored. In this article, we analyze a new pricing methodology for temperature derivatives that accounts for forward-looking information. More precisely, we provide an empirical back-up for the theoretical framework of so-called consistent factor models for temperature forecast curves introduced previously in the literature and put this pricing approach into practice. First, we perform a thorough statistical analysis of meteorological forecast curve data. Second, based on this analysis we propose a specific consistent two-factor model, derive explicit temperature derivative prices, and calibrate the market price of risk (MPR). The power of the model is demonstrated against alternative pricing models. This confirms that at least parts of the irregularity of the MPR observed in earlier studies are not due to irregular risk perception but rather due to information misspecification.