An approach to stochastic differential equations for long-term forecasting in the presence of $\alpha$-stable noise: an application to gold prices

Abstract

This article introduces a novel approach to forecasting gold prices over an extended period by leveraging a sophisticated stochastic process. Departing from traditional models, our proposed framework accommodates the non-Gaussian and non-homogeneous nature of gold market dynamics. Rooted in the $\alpha$-stable distribution, our model captures time-dependent characteristics and exhibits flexibility in handling the distinctive features observed in real gold prices. Building upon prior research, we present a comprehensive methodology for estimating time-dependent parameters and validate its efficacy through simulations. The results affirm the universality of our stochastic model, showcasing its applicability for accurate and robust long-term predictions in gold prices.