Replication Data and Code for “Supply Chain Readiness and Export Competitiveness in Africa: Evidence from Interpretable Machine Learning”

This dataset supports the study titled “Supply Chain Readiness and Export Competitiveness in Africa: Evidence from Interpretable Machine Learning.” The study examines whether supply-chain readiness predicts export competitiveness across African economies after removing Trade (% of GDP), an aggregate variable that includes exports and creates an accounting-overlap concern. The repository contains the processed country-year dataset and Python code used for the empirical analysis. The data cover 54 African economies from 1985 to 2024, producing a balanced panel of 2,160 observations. Export competitiveness is measured as exports of goods and services as a percentage of GDP. The modelling predictors include imports as a percentage of GDP, internet users, mobile subscriptions, electricity access, industry value added, services value added, foreign direct investment net inflows, population density, GDP per capita, and current account balance. Trade (% of GDP) is documented but excluded from the modelling feature set because it mechanically contains exports. The accompanying Python notebook reproduces the full analysis. It estimates linear regression, ridge regression, Random Forest, Extra Trees, and Gradient Boosting models. Model performance is assessed using random validation and temporal validation. The notebook reports mean absolute error, root mean squared error, and R-squared. It also generates permutation importance, feature response curves, subregional prediction errors, principal component analysis, and K-means clustering for export-readiness regime profiling. The files are shared to support transparency, replication, and further research on supply-chain readiness, export competitiveness, African subregions, temporal validation, and interpretable machine learning in development economics. The data were drawn from public development indicators and processed for the purpose of this study.