The AI for Trade Global Challenge
As geopolitical and economic uncertainties reshape international commerce, accurate predictions are crucial for businesses, policymakers, and researchers. The AI for trade global challenge invites data scientists, economists, and AI experts to explore the use of machine learning techniques to improve trade forecasting.
Goal
Teams will have to submit a forecast of the trade flows for the United States and China for the month of October of 2025. The forecast should provide trade values in dollars for the top 20 sources and destinations for the top 200 exports and imports of the United States and China (at the HS4 product level).
Key Dates
Submissions are due on Oct 15, 2025 and will be evaluated as soon as fine-grained trade data for the US and China for the month of October becomes available (usually about two months after (e.g. during December)).
Eligibility
Open to global participants (academics, industry professionals, students, and independent researchers).
Teams of 1-5 members (individual submissions allowed).
Multidisciplinary teams encouraged (AI/ML, economics, trade policy, data science).
Each participant must belong to a single team.
All participants must be able to identify themselves as natural persons (no pseudonyms or “sock puppet” accounts).
Teams must register to get access to the training data provided by the OEC.
Open Source
Teams are not required to open source their code, but are encouraged to do so. Teams must provide a two page description of their methods and must present their results in an online event as a condition for claiming the prize.
Resources
The Observatory of Economic Complexity (oec.world) will make available trade data for the United States and China for all months of the year 2023 and 2024. Teams can use this data, and any additional data, to create and validate their models.
Allowed Methods
Any forecasting method is allowed, general equilibrium models, regressions, neural networks, tarot cards, etc. Any method as long as it can be translated into a file like the one described in the submission guidelines.
Submission Guidelines
Teams must submit a plain text “.csv” file with their predictions midnight Oct 15, 2025 (Central European Time). The file must adhere to the following format:
“Country1” , “Country2”, “ProductCode”, “TradeFlow”, “Value”
With countries in ISO 3-letter codes, 4 digit HS codes for products, and trade flow forecasts in USD.
For example:
“USA”, “CHL”, “8404”, “Export”, “1234567”
“USA”, “CHN”, “8405”, “Import”, “1234567”
Together with their submission, teams must submit a two-page description of their method. Method descriptions go to the point, focusing on how the model was created and implemented (math is very much welcome). Ideally, a technical team should be able to reproduce the method using this document. The top three teams will be required to present their methods in a short presentation during the award ceremony (to be held online).
Evaluation
Predictions about the trade flows of the US and China will be evaluated respectively against trade data published by each of these countries (the US and China).
The winning team is the one with the best forecast according to sMAPE (symmetric mean absolute percentage error). To be eligible for winning, all forecasts must pass a minimum requirement of having a higher accuracy than a forecast based on using raw trade data for July 2025 to predict the flows for October 2025 (historical data shows these baseline sMAPEs to be about 35% to 45%).
Prizes
First Prize:
3000 USD
Free OEC Premium accounts for 1 year for all team members.
Second Prize:
2000 USD
Free OEC Premium accounts for 1 year for all team members.
Third Prize:
1000 USD
Free OEC Premium accounts for 1 year for all team members.
Partners
The AI for Trade Global Challenge is organized by the Center for Collective Learning in Partnership with the following organizations:
🌎 The Observatory of Economic Complexity (Strategic Data Partner)
🌎 Poverty and Equity at the World Bank
🌎 Trade Practice, The World Bank
🇪🇺 European Lighthouse of AI for Sustainability (ELIAS)
🇬🇧 The Supply Chain AI Lab at the University of Cambridge
🇬🇧 Complexity Economics Group Institute of New Economic Thinking (INET) Oxford University
🇭🇺 Corvinus Institute of Advanced Studies (CIAS) at Corvinus University of Budapest
🇫🇷 Institute for Advanced Study in Toulouse (IAST) at the Toulouse School of Economics
🇪🇸 Fundación Cotec, Spain
🇺🇸 Global Opportunity Lab at UC Berkeley