loreenach254/sub-saharan-africa-gdp-prediction

Machine learning project predicting GDP per capita in Sub-Saharan Africa using socioeconomic indicators. # Predicting GDP Per Capita in Sub-Saharan Africa Using Machine Learning ## Project Overview This project explores whether socioeconomic indicators can be used to predict GDP per capita across Sub-Saharan Africa using machine learning. The original objective was to investigate poverty prediction. However, during data exploration, poverty-related indicators contained substantial missing values, which limited the reliability of the modelling process. Instead of applying extensive imputation, the project direction was revised to predict GDP per capita, a widely used measure of economic development with better data availability. The project follows a complete end-to-end machine learning workflow: - Data preparation - Exploratory data analysis - Feature engineering - Model development - Model evaluation - Model interpretation - Docker-based reproducibility The main goal was not only to build a predictive model, but also to understand how socioeconomic conditions are associated with differences in economic development across countries. ## Research Question **Can socioeconomic indicators predict GDP per capita across Sub-Saharan African countries?** The predictors used in the analysis were: - Infant mortality rate - Rural population percentage - Clean water access - Unemployment rate - Year - Subregion The target variable was: - Log-transformed GDP per capita ## Dataset Description The final dataset contained: - 47 Sub-Saharan African countries - Data covering the period 1960–2024 - 3,099 observations before modelling - 2,794 observations used for modelling after removing missing GDP values The main variables included: | Variable | Description | |---|---| | gdp_per_capita | GDP per person | | infant_mortality | Infant mortality rate | | rural_population | Percentage of population living in rural areas | | clean_water_access | Percentage with access to clean water | | unemployment_rate | Unemployment percentage | | year | Observation year | | subregion | Re …