A Conditional Value Test of Exogenous Variables in Potato Price Forecasting
Public Health - Open Journal · 2026 · 11 (1) · 85-96
DOI: 10.17140/phoj.11.s1.85 ↗Kurz gefasst
Fachartikel in der Zeitschrift Public Health - Open Journal, 2026. Verfasst von Raj Shekar M und 5 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Whether weather variables improve agricultural price forecasts is usually answered on average, yet theoryand some evidence suggest that weather matters only in particular regimes, such as drought. This studyturns the question into one about conditional value: when does weather help, and can a rule detect thoseoccasions in real time, Using weekly potato prices with temperature and precipitation (January 2003 –September 2023), three forecasters are compared in an expanding-window walk-forward design coveringabout 15 years of out-of-sample forecasts: a naive no-change forecast (A), a price-only machine-learningmodel (B) and a model that adds lagged temperature and precipitation (C). A confidence gate, which opensonly when a model has shown consistent level-accuracy and turning-point improvements over its referenceon recently realised origins, is applied to B versus A and to C versus B, and raw prices are compared withKalman-filled prices as inputs. Price-only models clearly beat the naive forecast at four weeks ahead (relativeMAE 0.85-0.90, Diebold-Mariano p < 0.001) but not significantly at one week. Weather did not improveaccuracy on average at either horizon: relative to B, model C ranged from about 0.99 to 1.02 times B’s errorand was significantly worse for ridge regression at four weeks (p = 0.017). The gate between C and B stayedmostly closed (5-21% of origins), and its open and closed states were no better than chance at identifyingperiods in which weather later helped; the gate between B and A cost accuracy relative to always using B,because B was already reliably better than naive at four weeks. Descriptively, weather added value in thethree drought years in the evaluation sample (MAE ratio 0.97 at one week versus 1.03 in other years), but thecontrast was not statistically significant (p = 0.064). Kalman-filled inputs helped modestly for boosted treesat one week. The results support a conditional, regime-specific view of weather in these data but show thatdetecting the regime from recent forecast performance is hard, so a null result for the gate is informativeabout the limits of detecting conditional value.
Zitieren
APA
M, R. S., B, S. A. S., P, N. K., Jagadish, R., Kabber, B., & Nayak, H. (2026). A Conditional Value Test of Exogenous Variables in Potato Price Forecasting. Public Health - Open Journal, 11(1), 85-96. https://doi.org/10.17140/phoj.11.s1.85
Vancouver
M RS, B SAS, P NK, Jagadish R, Kabber B, Nayak H. A Conditional Value Test of Exogenous Variables in Potato Price Forecasting. Public Health - Open Journal. 2026;11(1):85-96. doi:10.17140/phoj.11.s1.85
BibTeX
@article{M2026,
author = {M, Raj Shekar and B, Sujith Arokia Swamy and P, Naveen Kumar and Jagadish, Ramya and Kabber, Bhavana and Nayak, Harish},
title = {A Conditional Value Test of Exogenous Variables in Potato Price Forecasting},
journal = {Public Health - Open Journal},
year = {2026},
volume = {11},
number = {1},
pages = {85--96},
doi = {10.17140/phoj.11.s1.85}
}
EUVIMED