A Horse Race Comparison of County-Level Crop Yield Prediction Methods

Junkan Li and Francis Tsiboe

No 391348, ARPC Brief from North Dakota State University

Abstract: Accurate county-level crop yield prediction is essential for agricultural outlooks and risk management, yet the predictive value of complex models remains uncertain. This study conducts a horse race comparison of alternative yield prediction methods for corn, soybeans, and cotton using USDA NASS yield data and PRISM weather data. Models range from simple historical averages to specifications incorporating spatial dependence, time dynamics, and weather variables. Evaluated using out-of-sample forecasts from 2015 to 2024, results show that simple models based on recent county-level yield averages consistently outperform more complex approaches. The findings highlight the robustness and practical value of parsimonious benchmarks for operational yield forecasting.

Keywords: Research Research Methods/Statistical Methods; Risk and Uncertainty (search for similar items in EconPapers)
Date: 2025-11-20
References: Add references at CitEc
Citations:

Downloads: (external link)
https://ageconsearch.umn.edu/record/391348/files/A ... 02025%E2%80%9314.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ags:arpcbr:391348

DOI: 10.22004/ag.econ.391348

Access Statistics for this paper

More papers in ARPC Brief from North Dakota State University
Bibliographic data for series maintained by AgEcon Search ().