A comprehensive method for medium-term analysis and forecasting (CMAF) of global monthly prices of agricultural commodities

Informazioni relative al progetto

CMAF

ID dell’accordo di sovvenzione: 101111405

DOI

10.3030/101111405

Progetto chiuso

Data della firma CE 26 Aprile 2023

Data di avvio 1 Giugno 2024

Data di completamento 31 Maggio 2026

Finanziato da

• Marie Skłodowska-Curie Actions (MSCA)

Costo totale

Nessun dato

Contributo UE

€ 183 600,96

Coordinato da

WIRTSCHAFTSUNIVERSITAT WIEN
Austria

Periodic Reporting for period 1 - CMAF (A comprehensive method for medium-term analysis and forecasting (CMAF) of global monthly prices of agricultural commodities)

Periodo di rendicontazione: 2024-06-01 al 2026-05-31

Sintesi del contesto e degli obiettivi generali del progetto

Work Performed and Main Achievements
The central scientific achievement of this project is the development of AGRICAF (Agricultural Commodity Forecasting Framework), a pipeline for the analysis and medium-term forecasting of global agricultural commodity prices. AGRICAF produces monthly price forecasts up to twelve months ahead using only publicly available data, and accompanies every forecast with explanations of the factors driving each prediction.
The framework operates in four stages. In the first stage, time series of potential explanatory variables are collected from publicly available sources, covering production figures, trade statistics, stock reports, and historical price data. In the second stage, a statistical screening process identifies which variables genuinely improve forecast accuracy for each commodity and time horizon, discarding those that add noise. In the third stage, a battery of ten statistical and machine learning models — including gradient boosting methods, random forests, linear and additive models, and time series decomposition techniques — is evaluated through cross-validation, with the best-performing model selected automatically for each combination of commodity, calendar month, and forecast horizon. In the fourth stage, every forecast is accompanied by explanations using model-agnostic interpretability methods, specifically Shapley values and LIME, organised across four levels of detail.
AGRICAF was developed and validated for five agricultural commodities: wheat, maize, soybean, arabica coffee, and sugar.
In parallel, the project produced a peer-reviewed scientific publication examining food prices under conflict conditions.

Lavoro eseguito dall’inizio del progetto fino alla fine del periodo coperto dalla relazione e principali risultati finora ottenuti

The central scientific achievement of this project is the development of AGRICAF (Agricultural Commodity Forecasting Framework), a complete open-source pipeline for the analysis and medium-term forecasting of global agricultural commodity prices.
The framework operates in four stages. In the first stage, time series of potential explanatory variables are collected from publicly available sources, covering production figures, trade statistics, stock reports, and historical price data. In the second stage, a statistical screening process identifies which variables genuinely improve forecast accuracy for each commodity and time horizon, discarding those that add noise. In the third stage, a battery of ten statistical and machine learning models — including gradient boosting methods, random forests, linear and additive models, and time series decomposition techniques — is evaluated through cross-validation, with the best-performing model selected automatically for each combination of commodity, calendar month, and forecast horizon. In the fourth stage, every forecast is accompanied by multi-level explanations using model-agnostic interpretability methods, specifically Shapley values and LIME, organised across four communication levels ranging from a broad annual overview down to a forensic decomposition of a specific forecast.
AGRICAF was successfully developed and validated for five globally traded agricultural commodities: wheat, maize, soybean, arabica coffee, and sugar — each requiring distinct datasets and independent model evaluation.
In parallel, the project produced a peer-reviewed scientific publication examining food prices under conflict conditions.

Progressi oltre lo stato dell’arte e potenziale impatto previsto (incluso l’impatto socioeconomico e le implicazioni sociali più ampie del progetto fino ad ora)

The project produced a working forecasting framework, validated across five agricultural commodity markets, with results documented on a public website and in an accompanying book. The forecasting performance shows that meaningful medium-term price forecasts can be built from publicly available data alone. A secondary finding, that the most useful predictors and best-performing models differ substantially depending on the commodity, forecast horizon, and time of year, is relevant to future research in this area.
At this stage, AGRICAF is a research tool. Translating it into something with real-world uptake would require several further steps: the methodology needs to be published in peer-reviewed form to establish scientific credibility; the framework would need to be extended to additional commodities, particularly those most critical to food security in low-income regions such as rice; and institutional partnerships with organisations such as FAO or IFPRI would be needed for the tool to reach the policymakers and analysts who could use it. Integration into existing monitoring platforms would be the most realistic pathway to broader use, as it would avoid asking end users to engage with a standalone tool they have no prior reason to consult.