Temperature Modeling in the Weather Derivative Pricing

This paper examines the theoretically obtained prices with values based on temperature data in the Isle of Man and the UK. We have also seen that the simulated temperature trajectories do not appear to include entire seasons where the temperature remains cooler than normal. Anecdotally we have all observed particularly mild winters or hot summers, and examination of the data reveals this to be more than just selective confirmation bias – there are certain winters that are persistently warmer than average throughout. The mean reverting nature of the model which appears to be a necessary component drags temperatures back towards average, reducing the possibility of such a season appearing in the modeled trajectories. It is possible that by incorporating long range weather forecasts, or by using a variable speed of mean reversion as applied in Zapranis and Alexandridis (2008) that this weakness can be overcome, but it does not seem clear from the literature weather there is a class of model which can overcome this problem. We have shown that the variabliiity of temperature is greatest during the winter periods, which suggests that where a company's results are correlated with temperature, the use of weather derivatives on the Isle of Man and the United Kingdom would be of most benefit during these periods