AI Stock Graphs for Theme Rotation and Price Attribution

Summary

This document describes a research framework for tracking how price moves may spread across companies linked by an AI theme. It builds a universe from equity contracts, excludes ETFs and indices, and uses a language model to assign companies to multiple sub-sector baskets with exposure and leader or follower roles. Search results and company news provide changing context for those baskets, while price data identifies unusual relative moves through within-basket Z-scores.

When a stock becomes an outlier, the system checks news for a possible explanation and uses statistical relationships to assess whether other basket members may follow. Signals are intended for notification and review before automated trading. The document gives implementation examples and describes dashboard monitoring, but reports no performance validation. It cautions that company classifications and news can be wrong, past lead-lag relationships may not persist, and live trading would need stronger controls for positions, sector exposure, losses, and execution.

Key ideas

• The stock universe should contain company equities and ADRs rather than ETFs, indices, or funds.
• Companies can have exposure to several thematic baskets, with different roles in each.
• Use current news to contextualize a basket, while using price anomalies to trigger investigation.
• A relative price deviation can identify a potential leader, but follower relationships require statistical validation.
• The framework is presented as research in notification mode, with further risk controls needed before automated trading.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.