Commodity Prices Volatility and Inflation in Ethiopia: Using Machine Learning Techniques

Commodity price fluctuations have significantly affected global and national economic stability, particularly in the wake of the COVID-19 pandemic and ongoing geopolitical tensions. High inflation, currently one of Ethiopia's most pressing economic challenges, has remained above 30% annually. This study examines how commodity price volatility on inflation in Ethiopia, using time-series data from 1992 to 2022. The research employs machine learning models, including XGBoost, Random Forest, and Gradient Boosting, as well as an Ordinary Least Squares (OLS) regression model to analyze both global and domestic shocks that influence inflation. The findings indicate that oil price growth, lagged domestic inflation, and global inflation are the most significant predictors of domestic inflation, underscoring Ethiopia's vulnerability to global commodity price changes and the persistence of inflationary trends. The Gradient Boosting model demonstrated superior predictive accuracy, explaining over 64% of the variation in inflation. The results suggest that effective policy measures focused on macroeconomic stability and the management of external shocks are crucial for mitigating inflationary pressures in Ethiopia. In conclusion, this research provides a novel approach that incorporates machine learning techniques to enhance understanding of inflation dynamics in developing economies, offering valuable insights for policymakers.