How Agent Selection Changes Order Flow in Shared-Model Markets

Summary

The study asks whether diversity among language model families persists in trading after agents are selected into submitting orders. In synthetic markets containing a fixed mixture of three model families, different presentations of news alter which families contribute orders. The reported financing and workforce-reduction events show that a family’s share of submitted orders can shift substantially, sometimes with little change in aggregate net orders and sometimes alongside a reversal in their direction. The analysis also considers homogeneous populations, where agents making decisions in the same direction leave less opposing flow.

An analytical decomposition explains how selection can change price accuracy even when aggregate demand sensitivity is unchanged. The results concern synthetic settings, particular presentation bundles, and submitted order flow; they do not establish how the effect generalizes to live markets. Cleaner replication and validation using a known-value setting are proposed as future work, so the evidence should be read as an initial result rather than a settled trading signal.

Key ideas

• News presentation can change which model families contribute submitted orders.
• Changes in family representation may occur without comparable changes in net order counts.
• Homogeneous decision populations can reduce opposing order flow.
• Selection can affect price accuracy even when aggregate demand sensitivity stays the same.
• The evidence is synthetic and prospective replication is still needed.

Tags

# Shared Models, Selective Trading, and Order Flow


# Shared Models, Selective Trading, and Order Flow









We study whether model diversity survives selection into trading. In synthetic markets with a fixed mixture of three language-model families, news presentation changes their representation among submitted orders. At the announcement round, Qwen's share of submitted orders shifts by 48 percentage points in the financing event, with little change in net order counts. In the workforce-reduction event, Mistral's share shifts by 40 percentage points while net counts reverse sign. Homogeneous populations remove opposing flow when their active decisions share a direction. An analytical decomposition shows why selection can improve or worsen price accuracy even at unchanged aggregate demand sensitivity. The evidence concerns presentation bundles and submitted flow; cleaner replication and a known-value validation are specified prospectively.

Shown in full with attribution under the source's licence. Licence: abstract CC0

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