Detecting Large Orders and Unusual Activity in the Order Book

Summary

The document raises the problem of identifying unusually large orders and possible informed or institutional activity from stock market order-book data. The author currently inspects plotted order sizes by eye, including cases where a large bid appears as prices fall, and asks for more systematic algorithms and research.

No detection method, paper, or empirical result is provided; the text is a question rather than a tutorial. It also does not establish that a large visible order comes from a single large trader or represents informed activity. Order size alone can be misleading because orders may be split, canceled, replenished, or spoofed, and interpretation requires context such as order flow and executions. The document therefore identifies a research problem but does not supply a solution or evidence for inferring smart money.

Key ideas

• The author seeks automated methods for detecting large orders in equity order-book data.
• The example relies on visual inspection of order size and price movement.
• A displayed large order does not by itself identify the trader or establish informed intent.
• The document provides no algorithms, cited studies, or empirical evidence.

Tags

# Known methods for big order detection


# Known methods for big order detection












I have an access to the order book from stock market and i am interested in finding an anomalous behaviour.

> What are the known methods, algorithms for detecting big orders or other activities of smart money?

I have built a program in python that plots a graph of all the order sizes from the order book. For instance when the price drops and i see a big buying order, then i know that this order is given by a big player. However i do this by eye and i would like to have a more accurate method. Are there any papers about this topic?

Thank you for help.

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.