American Options on Commodities Under Stochastic Convenience Yield and Stochastic Volatility

Abstract: American and Bermudan options have a wide range of applications in financial markets, e.g. in commodities markets among others. The pricing literature of such contingent claims is broad and many different algorithms and frameworks have been developed. The purpose of this thesis is to investigate how the Least-Squares Method (LSM), \cite{LSM}, can be extended to incorporate stochastic convenience yield and stochastic volatility in the pricing algorithm by using a commodity underlying. Moreover, the thesis aims to investigate the impact of stochastic convenience yield and stochastic volatility on the early exercise premium (EEP) of the American option written on a commodity. The results show that only the convenience yield increases the price of the American option. While, volatility does not add any edge to the algorithm when it is used as regressor. The insertion of the convenience yield increases the EEP especially for deep in the money options and long time span contracts. Lastly, the power polynomial specification shows better performances than the Laguerre one.