International agricultural trade forecasting using machine learning
Authors:
- Munisamy GopinathMGMunisamy Gopinath
- Feras A. BatarsehFeras A. Batarseh
- Jayson BeckmanJBJayson Beckman
- Ajay KulkarniAKAjay Kulkarni
- Sei JeongSJSei Jeong
Open Access
- 22 January 2021
- journal article
- research article
- Published by Cambridge University Press (CUP) in Data & Policy
- Vol. 3
Abstract
Focusing on seven major agricultural commodities with a long history of trade, this study employs data-driven analytics to decipher patterns of trade, namely using supervised machine learning (ML), as well as neural networks. The supervised ML and neural network techniques are trained on data until 2010 and 2014, respectively. Results show the high relevance of ML models to forecasting trade patterns in near- and long-term relative to traditional approaches, which are often subjective assessments or time-series projections. While supervised ML techniques quantified key economic factors underlying agricultural trade flows, neural network approaches provide better fits over the long term.Keywords
This publication has 21 references indexed in Scilit:
- Time differences, communication and trade: longitude matters IIReview of World Economics, 2013
- The prevention and handling of the missing dataKorean Journal of Anesthesiology, 2013
- Determinants of World Demand for U.S. Corn Seeds: The Role of Trade CostsAmerican Journal of Agricultural Economics, 2010
- Missing Data Analysis: Making It Work in the Real WorldAnnual Review of Psychology, 2009
- The Puzzling Persistence of the Distance Effect on Bilateral TradeReview of Economics and Statistics, 2008
- The Log of GravityReview of Economics and Statistics, 2006
- Gravity with Gravitas: A Solution to the Border PuzzleAmerican Economic Review, 2003
- Missing data: Our view of the state of the art.Psychological Methods, 2002
- Context-Sensitive Feature Selection for Lazy Learners1997
- Learning two-tiered descriptions of flexible concepts: The POSEIDON systemMachine Learning, 1992