Tracking Portfolio Optimization (TPFOpt) Studio

Tracking Portfolio Optimization (TPFOpt) Studio addresses the construction of an index-tracking portfolio from raw market data within a single Python workflow. The tool builds on my research in this area and is at an early stage of development.

The workflow is accessible through a web interface.

Key features

Data retrieval

Retrieve historical index and constituent data, including prices and compositions.

Preprocessing

Clean and align returns, and enrich them with fundamental and alternative data such as ESG scores and sector classifications.

Stock selection

Apply feature selection to identify the most relevant assets.

Portfolio optimization

Minimize tracking error subject to real-world constraints: UCITS (the 5/10/40 rule), cardinality, turnover, and sector exposure.

Performance and risk analysis

Evaluate tracking quality and risk exposure: tracking error, active share, turnover, and concentration measures.