This paper presents a novel method integrating Generalized Additive Models for Location, Scale, and Shape (GAMLSS) with Bayesian Markov-Switching GARCH (MSGARCH) models to enhance forecasting in commodity price returns, focusing on grain portfolios. We leverage GAMLSS to model non-normal distributions of return series, crucial for accurately simulating real options. These models then inform the Bayesian MSGARCH framework, improving projections of returns and volatility, essential for effective financial planning and risk management. This innovative approach not only advances practical portfolio management but also contributes to the theoretical development of real options theory. Demonstrating its efficacy, our methodology offers a more informed, strategic approach to the complex world of commodity trading, bridging the gap between theoretical models and practical financial applications.
Complete tool for submission, evaluation and publication of proceedings for scientific events.
