How Traders Infer Other Participants’ Order Intentions

Summary

The document asks whether strategic-trading models can reasonably assume that participants know one another’s intended orders. It describes a model in which all traders know the number of participants and each participant’s desired buy or sell quantity, allowing them to anticipate aggregate order flow. The response points to mean-field game research as a way to think about participants anticipating price pressure over longer horizons, including through published analyses of current flows and trends.

For shorter horizons, the response highlights liquidity imbalance as a signal that can help predict trading conditions and inform strategic order placement under adverse selection. The discussion offers references and examples of how traders may estimate others’ activity, but it does not establish that individual orders or intentions are generally observable. Its practical relevance depends on the market, available flow information, and signal quality; the answer does not quantify predictive performance or explain how to estimate the model’s parameters.

Key ideas

• Some strategic models assume traders know other participants’ intended order quantities and can anticipate net order flow.
• Mean-field game research models how participants anticipate price pressure over longer horizons.
• Intermediaries may publish flow and trend analyses that help market participants assess aggregate activity.
• Liquidity imbalance can serve as a short-horizon signal for strategic order placement.
• The discussion gives examples and references but does not show that individual intentions are directly observable or quantify signal accuracy.

Tags

# How do market participants know intentions of other players


# How do market participants know intentions of other players












I'm reading a paper on order anticipation strategy and came across this line in the paper:

> We assume that all model parameters, including `n` and `∆ := (∆0 , . . . , ∆n )`, are known to all strategic traders. Hence all traders anticipate the net orders of all other traders.

I have read similar arguments made in some other papers as well. I wonder if it is fair to make such an assumption. I havent read the entire paper because this line left me puzzled. Can someone please share some insights on this points.

Link to the paper

---EDIT---- In the above comment. n is the number of strategic participants and ∆i is the number of shares that the ith participants want to buy/sell.

## Answer by lehalle (score 6)

https://quant.stackexchange.com/a/33433

First of all, you may have a look at two other papers

- Mean-Field Game Strategies for Optimal Execution by Huang, Jaimungal and Nourian.

- Mean Field Game of Controls and An Application To Trade Crowding by Cardaliaguet and L

In the second one, we explain how on the long run, market participants try to anticipate price pressure. For instance intermediaries (ie investment banks and brokers) often publish analysis on the "current trends" in term of flows.

Moreover, on the short term, you can have a look at this other paper

- Limit Order Strategic Placement with Adverse Selection Risk and the Role of Latency by Mounjid and L.

We give in it the example of the predictive power of the liquidity imbalance. This "signal" is often used in intraday. At my knowledge, it has been popularized by Trade arrival dynamics and quote imbalance in a limit order book.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

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