📈 Finance

Financial experiments in Python — interactive notebooks demonstrating key concepts in quantitative finance, portfolio theory, and risk management, built with marimo.

📂 Project Structure

finance/
├── demos/                  # Interactive marimo notebooks
│   ├── prices_returns.py       Price series and return calculations
│   ├── normality.py            Distribution analysis of returns
│   ├── annualization.py        Annualizing returns and volatility
│   ├── rolling_window.py       Rolling statistics and windows
│   ├── sharp_ratio.py          Sharpe ratio analysis
│   ├── drawdown.py             Drawdown visualization
│   ├── duration_matching.py    Bond duration matching (immunization)
│   ├── value_at_risk.py        Value at Risk (VaR) estimation
│   ├── efficient_frontier1.py  Markowitz efficient frontier (2-asset)
│   ├── efficient_frontier2.py  Markowitz efficient frontier (n-asset)
│   ├── tracking_err.py         Tracking error and information ratio
│   ├── crash_correlations.py   Correlation breakdown during crashes
│   ├── candle_sticks.py        Candlestick charting
│   ├── valuation_multiples.py  Valuation multiples analysis
│   ├── cir.py                  CIR interest rate model
│   ├── bond_risk.py            Zero-coupon bond risk with CIR
│   └── cppi.py                 Constant Proportion Portfolio Insurance│   ├── cppi_liability.py        CPPI vs. alternatives for liability funding├── toolkit/                # Core library
│   ├── data.py                 Data fetching and preprocessing
│   ├── general.py              General-purpose utilities
│   ├── portfolio.py            Portfolio construction & optimization
│   ├── risk.py                 Risk measurement and analytics
│   └── ui.py                   Visualization helpers
├── research/               # 🔬 Personal market research (future)
├── tests/                  # Test suite
└── requirements.txt        # Dependencies

🚀 Getting Started

Prerequisites

• Python 3.10+
• pip

Installation

git clone <repo-url>
cd finance
pip install -r requirements.txt

Running the Notebooks

All demos are marimo notebooks. To run any demo:

marimo edit demos/<notebook>.py

Or start the marimo server to browse all notebooks:

marimo edit demos/

📊 Demos Overview

Notebook	Topic
prices_returns	Converting price series to returns (simple & log)
normality	Testing whether returns follow a normal distribution
annualization	Scaling daily statistics to annual equivalents
rolling_window	Computing rolling means, volatility, and correlations
sharp_ratio	Risk-adjusted return measurement
drawdown	Peak-to-trough decline analysis
duration_matching	Bond immunization via duration matching with CIR
value_at_risk	Parametric, historical, and Monte Carlo VaR
efficient_frontier1	Two-asset portfolio optimization
efficient_frontier2	N-asset Markowitz mean-variance optimization
tracking_err	Benchmark-relative risk metrics
crash_correlations	How correlations spike during market stress
candle_sticks	OHLC candlestick charting with volume
valuation_multiples	P/E, P/B, EV/EBITDA and other multiples
cir	Cox-Ingersoll-Ross interest rate model simulation
bond_risk	Zero-coupon bond risk analysis under CIR rates
cppi	Constant Proportion Portfolio Insurance strategy
cppi_liability	CPPI vs. alternative strategies for liability funding

🧪 Research

The research/ directory is reserved for my own market research — explorations, backtests, and empirical studies beyond textbook theory. This is where future work and findings will live.

🛠️ Toolkit

The toolkit/ package provides reusable building blocks:

• data — Fetch market data via yfinance, clean and transform series
• general — Math and statistical helpers
• portfolio — Portfolio weights, optimization, efficient frontier
• risk — VaR, CVaR, drawdowns, risk decompositions
• ui — Plotly and matplotlib-based visualization functions

📝 License

MIT © 2026 Szymon Wieloch — see LICENSE.