Weather Forecasting for Weather Derivatives

Abstract

We take a simple time-series approach to modeling and forecasting daily average temperature in U.S. cities, and we inquire systematically as to whether it may prove useful from the vantage point of participants in the weather derivatives market. The answer is, perhaps surprisingly, yes. Timeseries modeling reveals both strong conditional mean dynamics and conditional variance dynamics in daily average temperature, and it reveals sharp differences between the distribution of temperature and the distribution of temperature surprises. The approach can easily be used to produce not only short-horizon point forecasts, but also the long-horizon density forecasts of maximal relevance in weather derivatives contexts. We produce and evaluate both, with some success. We conclude that additional inquiry into nonstructural weather forecasting methods will likely prove useful in weather derivatives contexts.

Pricing weather derivatives: An application to the electrical utility

“…Applying the Campbell and Diebold (2002)'s time series model and Cao and Wei (2004)'s extended Lucas (1978) pure exchange economy model to the 32 years' temperature data in the city of Guangzhou, China, this paper derived the equilibrium prices for CDD forward, call and put options. The estimation of the market price of risk over a range of risk aversion parameters implies that the risk premium can be an important part of derivatives price and the market price of risk affects options much more than forwards.…”

“…According to Campbell and Diebold (2002), the daily temperature dynamics should include 367 three fundamental components: trends, seasonality, and cyclical volatility. The trend component reflects the global warming tendency, which is likely minor but relevant.…”

“…P and Q captures the seasonal cycle, and R captures the GARCH effects of the volatility. According to the minimum Akaike and Schwarz criteria, Campbell and Diebold (2005) suggested the adequate value for these parameters. The setting of L = 25, P = 3, Q = 3 and R = 1 has been employed.…”

“…The pricing mechanisms suggested in the literature (Alaton et al, 2002;Benth and Benth, 2005;Campbell and Diebold, 2002;Cao and Wei, 2004) are various, burn analysis, index modeling approaches, daily weather variable process models, and the so called daily simulation approaches to name a few. Burn analysis valuates a derivative simply by averaging all payoffs that would have been realized in the past.…”

Risk Management in Energy Markets

“…(See also Campbell and Diebold, 2002). In our model the deviation of the actual temperature from the historical average of temperature over the next day, t + 1, is a function of the deviation of the actual from the historical average of the temperature today, t, and the deviation of the actual from the historical average over the previous day, t − 1.…”