A portfolio optimization tool implementing Markowitz mean-variance optimization for constructing optimal portfolios. The tool uses Ledoit-Wolf shrinkage to improve covariance matrix estimation, addressing the instability of sample covariance matrices in high-dimensional settings. Built with Rust for efficient core optimization logic and Python/Streamlit for interactive visualization and analysis.
Install dependencies using pixi:
pixi installRun the Rust application:
pixi run startRun the Streamlit dashboard:
pixi run dash- Markowitz (1952). Portfolio selection
- Ledoit & Wolf (2004). A well-conditioned estimator for large-dimensional covariance matrices
- Fischer Black; Robert B Litterman (30 September 1991). "Asset Allocation: Combining Investor Views with Market Equilibrium". The Journal of Fixed Income. (not yet implemented)
NASDAQ listing: https://github.com/datasets/nasdaq-listings/tree/main/data