Metadata-Version: 2.4
Name: afrikana-risk
Version: 1.0.0
Summary: Production-ready quantitative risk toolkit — credit scoring, IFRS9 ECL, fraud detection, and model governance. Built in Africa, applicable anywhere.
Author-email: Peterson Mutegi <pitmuriuki@gmail.com>
License: MIT
Project-URL: Homepage, https://github.com/Peterson-Muriuki/afrikana-risk
Project-URL: Repository, https://github.com/Peterson-Muriuki/afrikana-risk
Project-URL: Documentation, https://github.com/Peterson-Muriuki/afrikana-risk#readme
Project-URL: Bug Tracker, https://github.com/Peterson-Muriuki/afrikana-risk/issues
Keywords: credit-risk,credit-scoring,ifrs9,basel,ecl,fraud-detection,model-governance,scorecard,pd-lgd-ead,stress-testing,champion-challenger,africa,fintech,banking,quantitative-finance,machine-learning,risk-analytics
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Financial and Insurance Industry
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Office/Business :: Financial
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10
Description-Content-Type: text/markdown
Requires-Dist: numpy>=1.24
Requires-Dist: pandas>=2.0
Requires-Dist: scikit-learn>=1.3
Requires-Dist: scipy>=1.11
Requires-Dist: lifelines>=0.27
Requires-Dist: statsmodels>=0.14
Requires-Dist: lightgbm>=4.0
Requires-Dist: optuna>=3.3
Provides-Extra: fraud
Requires-Dist: tensorflow>=2.13; extra == "fraud"
Requires-Dist: torch>=2.0; extra == "fraud"
Provides-Extra: ai
Requires-Dist: mistralai>=0.4; extra == "ai"
Provides-Extra: viz
Requires-Dist: matplotlib>=3.7; extra == "viz"
Requires-Dist: seaborn>=0.12; extra == "viz"
Requires-Dist: plotly>=5.15; extra == "viz"
Provides-Extra: dev
Requires-Dist: pytest>=7.4; extra == "dev"
Requires-Dist: pytest-cov>=4.1; extra == "dev"
Requires-Dist: black>=23.0; extra == "dev"
Requires-Dist: ruff>=0.1; extra == "dev"
Requires-Dist: mypy>=1.5; extra == "dev"
Requires-Dist: pre-commit>=3.4; extra == "dev"
Provides-Extra: all
Requires-Dist: afrikana-risk[ai,dev,fraud,viz]; extra == "all"

# afrikana-risk

Production-ready quantitative risk toolkit for credit scoring, IFRS9 ECL, fraud detection, and model governance.

Built in Africa. Applicable anywhere.

---

## Features

### Credit Risk
- PD / LGD / EAD modeling
- Scorecard building (WoE, IV)
- Basel III capital estimation

### IFRS 9
- Stage 1 / 2 / 3 ECL
- Macro-scenario overlays
- Probability-weighted ECL

### Stress Testing
- Scenario analysis (base / adverse / severe)
- Credit VaR (Monte Carlo, Vasicek)
- NPL trajectory projection

### Fraud Detection
- Isolation Forest anomaly detection
- Supervised fraud classification
- Rule-based risk flags
- Ensemble scoring

### Model Monitoring
- PSI (Population Stability Index)
- Gini / KS tracking
- Feature drift detection

### Model Governance
- Champion-Challenger framework
- DeLong AUC significance testing
- Audit trail for regulatory compliance

---

## Installation

```bash
pip install afrikana-risk
