Can the Topological Shape of the Market Predict How Returns Evolve?

I'm pleased to share that my paper is now available on both SSRN and arXiv.

The core idea: I ask how strangely stocks are positioned within the cross-sectional point cloud at any given moment. Using BallMapper to build dynamic topological graphs, a decoder-conditional VAE to score structural surprise, and Function-on-Function regression to link those scores to future intraday returns — the paper shows that topology carries genuine predictive signal.

Cross-Sectional Topological Anomaly Scores and Intraday Return Predictability: BallMapper Descriptors, Decoder-Conditional VAE, and Function-on-Function Regression on S&P 500 Intraday Data

• 📄 (SSRN): https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6837799
• 📄 (arXiv): https://arxiv.org/abs/2606.08586