Fractional Long-Memory Dynamics and Residual Machine Learning for Medium-Horizon Agricultural Commodity Price Forecasting
Fractal and Fractional · 06.08.2026 · 10 (8) · 536
DOI: 10.3390/fractalfract10080536 ↗Kurz gefasst
Fachartikel in der Zeitschrift Fractal and Fractional, online erschienen am 06.08.2026. Verfasst von Sergio Orozco Cirilo und 7 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Universidad de Guanajuato.
Abstract (Originalsprache)
This paper develops a hybrid fractional-order framework for medium-horizon forecasting of wheat, corn, and soybean prices, combining a Caputo fractional differential equation with exogenous macro-climatic drivers (weather, crude oil, exchange rate, inflation) and a machine learning residual-correction layer. Existence, uniqueness, and Ulam–Hyers stability are established for both Caputo and Atangana–Baleanu formulations via Banach fixed-point theory, with numerical illustration through a fractional Adams–Bashforth–Moulton predictor–corrector scheme. Formal unit-root tests (ADF, KPSS, Phillips–Perron) confirm I(1) behaviour in log-price levels and stationarity in first differences. Residuals are corrected using XGBoost and two feedforward neural networks (MLP-A, MLP-B), producing a family of hybrid forecasting models. Using 25 years of monthly FRED data (January 2000–December 2024), multi-method long-memory diagnostics (Hurst exponent, Lo’s modified R/S test, DFA, local Whittle estimation) confirm near unit-root fractional integration in log-price levels (H∈[0.985,1.044]), with estimated fractional orders α^∈{0.737,0.884,0.451} for wheat, corn, and soybean. Under a strict rolling-origin protocol (17 origins, horizons h∈{1,5,6,12} months), conventional benchmarks remain competitive at h=1, but their MAPE degrades to 16.6–31.9% at h=12, while Frac+XGBoost error stays flat at 3.9–14.5%. This horizon-robust advantage holds across three market regimes (COVID-19 pandemic shock, 2021–2022 super-cycle, 2023–2024 normalisation) and is confirmed by Holm–Bonferroni-corrected Diebold–Mariano tests (p
Förderung
- Universidad de Guanajuato
Zitieren
APA
Cirilo, S. O., Vargas-Canales, J. M., Jarquín, D. M. S., Cuéllar, S. E. M., Bautista, J. A., Cobos, A. V., Haros, B. R., & Hernández, B. H. (2026). Fractional Long-Memory Dynamics and Residual Machine Learning for Medium-Horizon Agricultural Commodity Price Forecasting. Fractal and Fractional, 10(8), 536. https://doi.org/10.3390/fractalfract10080536
Vancouver
Cirilo SO, Vargas-Canales JM, Jarquín DMS, Cuéllar SEM, Bautista JA, Cobos AV, et al. Fractional Long-Memory Dynamics and Residual Machine Learning for Medium-Horizon Agricultural Commodity Price Forecasting. Fractal and Fractional. 2026;10(8):536. doi:10.3390/fractalfract10080536
BibTeX
@article{Cirilo2026,
author = {Cirilo, Sergio Orozco and Vargas-Canales, Juan Manuel and Jarquín, Dora María Sangerman and Cuéllar, Sergio Ernesto Medina and Bautista, Juan Antonio and Cobos, Alberto Valdes and Haros, Benito Rodríguez and Hernández, Belén Hernández},
title = {Fractional Long-Memory Dynamics and Residual Machine Learning for Medium-Horizon Agricultural Commodity Price Forecasting},
journal = {Fractal and Fractional},
year = {2026},
volume = {10},
number = {8},
pages = {536},
doi = {10.3390/fractalfract10080536}
}
EUVIMED