Pricing Of Weather Derivatives Based On The Time Series

In this paper,an accurate pricing method has been established for the weather derivatives market.Our main work is to analyze the temperature's characteristic,such as,trend,seasonal and heteroscedasticity of the temperature data by time series analysis method,so as to establish a stochastic model to fit the temperature change law.Based on the analysis of the temperature data,it is concluded that the temperature model has a linear trend,periodic,seasonal and heteroskedasticity.A stochastic model with GARCH mean regression Ornstein-Uhlenbeck is established for the historical temperature of 62 years.The volatility is seasonal and weak trend improvement model,through the temperature model to obtain a future period of the temperature forecast.After the forecast value is obtained,the derivative pricing model is established by the no-arbitrage pricing theory.The Monte Carlo method can be used to simulate the price of the derivative.Through the analysis of the actual value and the predicted value obtained from the stochastic model in 2013,the validity of the temperature model is verified by the relative error.In the model used to predict the future weather value,only the historical data can be used.For the meteorological information and the expert analysis index,the pricing formula of the weather derivatives can be deduced by the uncertainty theory,and the pricing method under the new angle can be realized.