This thesis focuses on the modeling of the (yearly) seasonality in the volatility of agricultural futures markets. The Samuelson Effect is incorporated in the futures prices using the proposed model by [ST17]. Furthermore, it defines the variance process as a meanreverting stochastic process with seasonal long run average mean ∅. As input for ∅, it proposes five different types of seasonality functions as well as a constant one. Additionally, it derives a state-space representation of the proposed models and estimates the parameters using the Kalman filter algorithm discussed in [Har89] and [Tsa10]. This thesis also provides an in-depth analysis of three agricultural commodities namely corn, soybeans and cotton. For all three commodities, the study uses time series data of futures contracts with fixed maturities and expands the estimation process to a data set containing constant maturity futures as well. Finally, it compares the results of the estimation for each data set and proposes a model for each of the three commodities considered. Overall, the study concludes that by choosing a seasonal stochastic volatility model in agricultural futures markets, this indeed increases the accuracy in the modeling. On the other hand, this thesis shows that by using a data set consisting of constant maturity futures, the seasonal pattern in the variance is much better captured and modeled.     «

This thesis focuses on the modeling of the (yearly) seasonality in the volatility of agricultural futures markets. The Samuelson Effect is incorporated in the futures prices using the proposed model by [ST17]. Furthermore, it defines the variance process as a meanreverting stochastic process with seasonal long run average mean ∅. As input for ∅, it proposes five different types of seasonality functions as well as a constant one. Additionally, it derives a state-space representation of the propos...     »