LLMs for Equities Feature Forecasting at Two Sigma with Ben Wellington - #736

This article from Practical AI discusses Two Sigma's use of Large Language Models (LLMs) in equities feature forecasting. It highlights the end-to-end process, from feature identification and data collection to model building and market behavior prediction. The article emphasizes the importance of a platform-centric approach, the impact of multimodal LLMs, strict data timestamping, and build-versus-buy decisions. It also touches upon the use of open-source models and the future of agentic AI in quantitative finance. The interview with Ben Wellington provides valuable insights into the practical application of AI in the financial industry.

• •Two Sigma uses LLMs for equities feature forecasting.
• •The process involves feature identification, data collection, model building, and market prediction.
• •Key considerations include platform-centric approach, data timestamping, and build vs. buy decisions.