Five Jupyter notebooks, a reusable Python module, and a complete PDF guide / ebook — everything you need to implement professional market risk models.
A complete, structured Python implementation of market risk — not scattered snippets, but a cohesive, professional pack.
Step-by-step walkthroughs for returns & volatility, EWMA & GARCH, VaR models, Expected Shortfall, and VaR backtesting. Run them top to bottom and understand every line.
A clean market_risk/ package with data.py, volatility.py, var.py, expected_shortfall.py, and backtesting.py. Import into your own projects immediately.
A full interactive risk dashboard (streamlit_app.py) for exploring any ticker — VaR, ES, GARCH volatility, and backtesting results in a live UI.
A teaching manual with full LaTeX equations, implementation notes, model interpretation, and risk manager language. Covers every concept in the notebooks.
A full set of exercises with worked solutions, suitable for self-study or classroom use. Covers all five topic areas with analytical and implementation questions.
A pinned requirements.txt and README with a one-command install and core usage examples. Run the dashboard or notebooks in minutes.
Each notebook builds on the previous one — from raw price data to a professionally backtested risk model.
Start from raw prices. Compute daily simple returns, convert to dollar losses, and estimate rolling volatility. Understand why volatility clustering makes constant-vol models dangerous.
Move beyond rolling windows. Implement RiskMetrics EWMA (λ=0.94) and fit a GARCH(1,1) model. Forecast conditional volatility forward and compare model behavior during stress periods.
Implement and compare four VaR approaches on a real portfolio. Understand the assumptions behind each model and when each one breaks down — the core of any practical risk workflow.
Go beyond VaR. Compute Expected Shortfall (CVaR) — the average loss in the worst α% of days. Understand why regulators and institutions now prefer ES over VaR for tail risk measurement.
A VaR model that can't be validated is just a guess. Implement exception counting, Kupiec unconditional coverage test, Christoffersen independence test, and a Basel traffic-light summary.
Whether you're studying for the FRM, working in risk, or building Python tools — this pack gives you exactly what you need.
Learn the theory and the implementation together — not one without the other. Exercises and answers included.
A clean, reusable Python module you can adapt for internal workflows. EWMA, GARCH, four VaR models, ES, and Basel backtesting — production-quality implementations.
A well-structured reference codebase for financial risk modeling. Clean module architecture, typed functions, and a Streamlit app you can extend or deploy immediately.
The full toolkit for daily market risk measurement — from volatility estimation to regulatory backtesting.
21-day rolling vol and RiskMetrics EWMA with configurable lambda. Annualized output, clustering diagnostics, and side-by-side comparison.
Full GARCH estimation via the arch library, conditional volatility extraction, and multi-day forecasting. Fits to any equity, crypto, or FX returns series.
Historical simulation, normal-parametric, EWMA-parametric, and GARCH-parametric VaR. Compare models on the same portfolio in a single function call.
Historical and parametric ES — the coherent tail risk measure now required under Basel III. Always paired with VaR so you see both metrics together.
Exception counting, Kupiec LR test, Christoffersen independence test, and Basel traffic-light zones. Know whether your model actually works.
No subscriptions. No locked features. Pay once, own it forever.
market_risk/ Python module (5 files)Five notebooks. A reusable module. A live dashboard. A complete PDF guide / ebook. Everything to go from zero to a production-quality market risk workflow in Python.