Moggycool/Forecasting-Financial-Inclusion-in-Ethiopia

Build a forecasting system that tracks Ethiopia's digital financial transformation using time series methods # Forecasting Financial Inclusion in Ethiopia (Unified Dataset + Enrichment + EDA) This repository contains a reproducible pipeline to: 1) validate and enrich a unified Financial Inclusion (FI) dataset for Ethiopia, and 2) generate EDA tables used to support narrative insights and forecasting readiness. The workflow is designed to be grader/reviewer-friendly: - deterministic CLI commands - explicit schema expectations - diagnostics outputs when validation issues are found - written insights + limitations in `INSIGHTS.md` --- ## Repository Structure (key items) - `data/raw/ethiopia_fi_unified_data.csv` Unified FI dataset (raw/un-enriched input) - `data/enrichment/new_records.yaml` Human-authored enrichment records (events, impact links, targets, etc.) - `data/processed/ethiopia_fi_unified_data__enriched.csv` Output of enrichment pipeline - `data/processed/diagnostics/` Validation diagnostics emitted by enrichment pipeline (only written when issues exist) - `scripts/apply_enrichment.py` Applies YAML enrichment records and runs relationship diagnostics - `scripts/run_exploration.py` Runs EDA tables (counts, temporal range, coverage, events, links) - `src/fi/` Core library modules: `io`, `validation`, `enrich`, `explore` - `INSIGHTS.md` Written insights using a Claim/Evidence/Interpretation/Confidence structure + limitations --- ## Quickstart ### 1) Create environment (example) Use your preferred environment manager. Example with `venv`: ```bash python -m venv .venv # Windows: .venv\Scripts\activate # macOS/Linux: source .venv/bin/activate pip install -r requirements.txt ``` ## Project Tree ```text Forecasting-Financial-Inclusion-in-Ethiopia ├─ dashboard │ └─ app.py ├─ data │ ├─ enrichment │ │ └─ new_records.yaml │ └─ processed │ ├─ diagnostics │ ├─ eda │ ├─ eda_enriched │ │ ├─ counts__category.csv │ │ ├─ counts__pillar.csv │ │ ├─ counts__record_type.csv │ │ ├─ events.csv │ │ ├─ impact_links.csv │ │ ├ …