Journal of Modelling in Management, Journal Year: 2025, Volume and Issue: unknown
Published: April 24, 2025
Purpose Investors and governments have traditionally depended on estimates of prices various commodities. Using data from 08/23/2013 to 04/15/2021, this study aims investigate the challenging problem forecasting scrap steel that are published a daily basis for east China regional market. The research has not given much attention predictions important commodity price indicator. Design/methodology/approach Gaussian process regression models, which estimated using cross-validation approaches with Bayesian optimizations, used provide forecasts. Findings Having relative root mean square error 0.4357%, constructed models appropriately generate forecasts out-of-sample testing stage 09/17/2019 04/15/2021. Originality/value Models designed can be by investors make well-informed decisions.
Language: Английский