Browsing by Author "Evarest, Emmanuel"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Modelling and pricing of weather derivatives: a regime switching approach
    (University of Dar es Salaam, 2018) Evarest, Emmanuel
    In this thesis, regime switching models for predicting temperature dynamics that is used for pricing weather derivatives are presented. Using historical temperature data from Swedish Meteorological and Hydrological Institute (SMHI) in Sweden and Tanzania Meteorological Agency (TMA) in Tanzania, the performance of the models are tested by evaluating their accumulated heating degree days, cooling degree days and cumulative average temperature. The test performance results show that, the regime switching model with heteroscedasticity is relatively better compared to other models, and it is used for pricing weather derivative futures and options. Then pricing of weather derivatives written on temperature indices is presented, where mathematical expressions for pricing weather futures and Monte Carlo simulations for pricing weather option contracts are presented. Also, Sensitivity analysis of weather derivative option prices with respect to model parameters is performed and the results show significant change in the prices of weather option contract for different changes in model parameters point estimates. Furthermore, it is shown that the regime switching models with Brownian motion and with an Ornstein-Uhlenbeck process as their shifted regimes respectively, have equal predictive accuracy levels, but weather derivative option contracts based on these models have different prices.
  • No Thumbnail Available
    Item
    Pricing of energy by means of stochastic model
    (Unversity of Dar es Salaam, 2008) Evarest, Emmanuel
    In this dissertation we present a mean-reverting jump diffusion model for pricing of energy particularly electricity by means of stochastic. We discuss the stochastic model which is used to model the behavior of electricity prices. Despite some distributional similarities with asset prices, electricity prices have dramatically different stochastic properties from those financial products even other commodities due to its non-storability nature. These properties include mean-reversion, stochastic volatility, seasonality as well as short lived spikes or jumps. The recent deregulation of electricity markets in the world has exposed power producers and users to market risk due to those unique features of energy price dynamics. The prices contain strong mean reversion, which reflects the demand and supply movements. The model developed is calibrated using the market data from Nord pool for the period from January 1997 to April 2000. The daily price model is estimated via maximum Likelihood-Conditional Characteristic Function (ML-CCF) to obtain the solution in closed form. From the model we derive the corresponding forward prices under Q-martingale measure and calculate forward prices at different expiries. All forward prices are subject to the market price of risk due to the fact that power markets are incomplete markets. The ability to model the spot prices and obtain forward price dynamics is essential when assessing the performance of heading strategies that use forward contacts.