Temperature Weather Derivatives Modeling And Pricing Under Regime Switching

Weather derivative,with its underling including a variety of weather indexes,is a new financial instrument to manage weather risks.Based on the daily average temperature residuals data in Zhengzhou,four regime-switching models are established and compared to determine the most accurate model to describe temperature data in Zhengzhou.The parameters of Gaussian mixture models are estimated by the expectation-maximization(EM)algorithm,and then four models are used to predict the daily average temperature in Zhengzhou from 1/1/2021 to 12/31/2021.Comparisons with the real temperature data indicate that the model with constant exponential time-varying mean-reverting and Brownian motion is the best effective regime-switching model to describe the stochastic dynamics of the temperature in Zhengzhou.For the model with a constant exponential time-varying mean-reverting and a Brownian motion,we calculate the predicted cumulative index values,the real cumulative index values,and the relative error rates with different base temperatures.It is shown that the proposed model has high accuracy and good stability,and can be used to price temperature options.Finally,we price temperature option with Monte Carlo simulations,and compare the option prices for different number of Monte Carlo simulations.The empirical results reflect the diversity of global climate and give the government advices for managing weather risk efficiently.It is also shown that we need to analyze the regional temperature data to construct the temperature model which is accurate to the local regional temperature characteristics,so as to more accurately price weather derivatives.