weldie10/forecasting-ethiopia-financial-inclusion

Time-series forecasting and event impact analysis of financial inclusion indicators in Ethiopia. Uses historical data, policy events, and machine learning to model trends, assess policy effects, and generate evidence-based insights for decision-making. # Ethiopia Financial Inclusion Forecasting Project for forecasting financial inclusion in Ethiopia using comprehensive data exploration, analysis, modeling, and interactive visualization. ## Project Structure ``` ├── .github/workflows/ │ └── unittests.yml # CI/CD workflow for unit tests ├── data/ │ ├── raw/ # Starter dataset │ │ ├── ethiopia_fi_unified_data.xlsx │ │ └── reference_codes.xlsx │ └── processed/ # Analysis-ready data ├── notebooks/ │ ├── 01_data_exploration.ipynb # Task 1: Data exploration │ ├── 02_data_enrichment.ipynb # Task 1: Data enrichment │ ├── 03_eda.ipynb # Task 2: Exploratory Data Analysis │ ├── 04_impact_modeling.ipynb # Task 3: Event Impact Modeling │ └── 05_forecasting.ipynb # Task 4: Forecasting Access and Usage ├── src/ │ ├── __init__.py │ ├── task1_data_exploration.py # Task 1: OOP data exploration & enrichment │ ├── task2_eda.py # Task 2: OOP exploratory data analysis │ ├── task3_impact_modeling.py # Task 3: OOP event impact modeling │ ├── task4_forecasting.py # Task 4: OOP forecasting module │ └── dashboard_components.py # Task 5: Dashboard components ├── dashboard/ │ └── app.py # Streamlit dashboard application ├── tests/ │ └── __init__.py ├── models/ # Trained models ├── reports/ │ └── figures/ # Generated visualizations ├── requirements.txt ├── README.md └── .gitignore ``` ## Setup 1. Create a virtual environment: ```bash python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate ``` 2. Install dependencies: ```bash pip install -r requirements.txt ``` ## Usage ### Task 5: Dashboard Development (Main Interface) Interactive Streamlit dashboard for exploring data, understanding event impacts, and viewing forecasts. #### Quick Start ```bash # Install dependencies (if not already installed) pip …