melimhur-eng/ethiopia-fi-forecast

# Ethiopia Financial Inclusion Forecast A forecasting system for Ethiopia's digital financial transformation, built for Selam Analytics (10 Academy KAIM 9 — Week 11 challenge). The system tracks and forecasts the two core Global Findex dimensions of financial inclusion: - **Access** — Account Ownership Rate - **Usage** — Digital Payment Adoption Rate It combines a sparse historical record (5 Findex survey waves, 2011–2024) with a catalog of policy/product/infrastructure **events** and modeled **event → indicator impact links**, to produce trend, event-augmented, and scenario forecasts through 2027. ## Project Structure ``` ethiopia-fi-forecast/ ├── .github/workflows/unittests.yml # CI: installs deps, runs pytest ├── data/ │ ├── raw/ # Starter dataset (as provided) + reference codes │ └── processed/ # Enriched / analysis-ready data (Task 1 output) ├── notebooks/ # Task 1-4 notebooks ├── src/ # Reusable data/modeling/forecasting code ├── dashboard/app.py # Streamlit dashboard (Task 5) ├── tests/ # Unit tests for src/ ├── models/ # Saved model artifacts / forecast tables ├── reports/figures/ # Exported charts used in reports ├── requirements.txt └── README.md ``` ## Setup ```bash python3 -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate pip install -r requirements.txt ``` ## Running the dashboard ```bash source .venv/bin/activate streamlit run dashboard/app.py ``` ## Running tests ```bash source .venv/bin/activate pytest -q ``` ## Data The unified dataset (`data/raw/ethiopia_fi_unified_data.csv`) uses a single schema for four record types, distinguished by `record_type`: | record_type | `category` | `pillar` | Meaning | |---|---|---|---| | `observation` | empty | set | Measured value from a survey/operator/regulator source | | `event` | …