Connectedness Portfolio

This project analyzes the connectedness of commodity markets and performs portfolio optimization using both covariance and DYIC (Dynamic Connectedness Index) matrices.

Usefull links

• https://gabauerdavid.github.io/ConnectednessApproach/
• https://www.financialconnectedness.org/

Features

• Fetches and cleans historical commodity data from Yahoo Finance using yfinance
• Computes rolling-window connectedness measures (DYIC) using VAR models
• Visualizes connectedness and portfolio performance
• Portfolio optimization using [PyPortfolioOpt] with both covariance and DYIC matrices
• Compares optimized portfolios with an equally weighted benchmark

Project Structure

• code.ipynb: Main Jupyter notebook containing all code, analysis, and visualizations
• pyproject.toml: Project dependencies and configuration (uses Poetry)
• poetry.lock: Locked dependency versions

Requirements

• Python >= 3.13
• See dependencies in pyproject.toml

Getting Started

1. Install dependencies
   Using Poetry:

   poetry install

   Using pip:

   pip install -r requirements.txt

2. Run the notebook
   Open code.ipynb in VS Code or JupyterLab and run all cells.

Main Packages Used

• numpy, pandas: Data manipulation
• yfinance: Downloading financial data
• matplotlib, seaborn: Visualization
• statsmodels: VAR modeling and connectedness analysis
• pyportfolioopt: Portfolio optimization

Author

Etienne Larchet
MIT License

---