News

Researchers rely on machine learning to predict economic trends. But what about ancient wisdom? A new paper by WorldQuant University student Alhassan Iddrisu and two co-authors combines artificial intelligence with an ancient African system of divination to explain currency trends in Ghana.

Their paper, published in the International Journal of Research and Scientific Innovation, explores the relationship between Ghana’s cedi and fuel prices, using modern data combined with cultural factors. The study finds that when fuel prices go up, the cedi tends to strengthen rather than weaken, going against common economic expectations.

The authors cite several possible factors: Government fuel subsidies might have softened the blow of global oil price hikes on everyday costs; increased tourism could have brought more foreign currency, easing pressure on the cedi; strict monetary policies likely attracted overseas investments, helping stabilize the exchange rate; and positive investor confidence in Ghana's debt restructuring efforts might have contributed to currency resilience.

These are predictable data points for machine learning models. What sets the new paper apart is its hybrid method for forecasting. On one side, the researchers use machine learning to make precise, testable predictions about currency and fuel trends. On the other, the researchers incorporate geomancy, an ancient African system of divination involving symbolic interpretations of patterns in the earth or sand.

Geomancy adds a layer of cultural insight, valuable in societies where beliefs and traditions influence economic decisions. By combining ancient and modern thinking, the researchers argue for a more holistic understanding of complex systems, where data meets storytelling and heritage.

The study offers practical advice for different groups. Policymakers should keep up disciplined financial strategies to balance currency strength with inflation control, phase in fuel price changes with protections for the poor, and use cedi stability to draw investments in areas like renewable energy and agriculture. Investors and analysts are encouraged to adopt similar mixed models for better risk evaluation, factoring in cultural elements when assessing African markets.

The study recommends further research to test this framework on other currencies, integrate it with various traditional systems worldwide, and push for research that respects non-Western knowledge.

Indigenous wisdom can complement scientific tools, challenging purely Western approaches to economics. The paper expands how analysts think about forecasting, making it more inclusive and interpretive. Ultimately, it calls for economics models that harmonize logic with legacy, inviting a future where AI illuminates rather than replaces ancestral insights.

Read more
Enoch Deyaka Mwini, Alhassan Iddrisu, Alfred Asiwome Adu (2025). Cedi Appreciation Relative to Fuel Prices: A Machine Learning and Ancient Geomantic Approach. Journal of Research and Scientific Innovation, 12(10).

About the WQU author
Alhassan Iddrisu works at Ghana’s Ministry of Finance. Before enrolling in the Master of Science in Financial Engineering program at WorldQuant University, he earned an undergraduate degree in statistics at the University for Development Studies in Ghana.