Paper: arXiv 2403.13429
Authors: Kaushalya Kularatnam, Tania Stathaki
Abstract
As algorithmic trading and electronic markets continue to transform the landscape of financial markets, detecting and deterring rogue agents to maintain a fair and efficient marketplace is crucial. The explosion of large datasets and the continually changing tricks of the trade make it difficult to adapt to new market conditions and detect bad actors. To that end, we propose a framework that can be adapted easily to various problems in the space of detecting market manipulation. Our approach entails initially employing a labelling algorithm which we use to create a training set to learn a weakly supervised model to identify potentially suspicious sequences of order book states. The main goal here is to learn a representation of the order book that can be used to easily compare future events. Subsequently, we posit the incorporation of expert assessment to scrutinize specific flagged order book states. In the event of an expert’s unavailability, recourse is taken to the application of a more complex algorithm on the identified suspicious order book states. We then conduct a similarity search between any new representation of the order book against the expert labelled representations to rank the results of the weak learner. We show some preliminary results that are promising to explore further in this direction
Complexity vs Empirical Score
- Math Complexity: 6.5/10
- Empirical Rigor: 4.0/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper employs advanced machine learning techniques like Temporal Convolutional Networks with dilations and specific activation functions (swish), showing substantial math complexity. However, the empirical section presents only preliminary results with small datasets and no live backtesting or robust statistical validation, placing it in the Lab Rats quadrant.
Research Flowchart
flowchart TD
A["Research Goal: Detect and triage<br>market manipulation in order books"] --> B["Data: Historical order book sequences"]
B --> C["Method: Weakly Supervised<br>Temporal Convolutional Network"]
C --> D{"Model Output<br>Identifies suspicious sequences"}
D --> E{"Expert Assessment Available?"}
E -- Yes --> F["Incorporate Expert Labels<br>Refine Model"]
E -- No --> G["Apply Complex Algorithm<br>Filter & Rank Alerts"]
F --> H["Outcomes: Framework for<br>Adaptive Market Surveillance"]
G --> H
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes a framework that uses a weakly supervised temporal convolutional network to learn representations of order‑book states for detecting and ranking spoofing, with expert labeling and similarity search for ranking.
Source sentence
we propose a framework that can be adapted easily to various problems in the space of detecting market manipulation.
Assumes 1
- The model is causal, i.e., predictions at time t depend only on data up to t and not on future information. model
Source sentence
Our model is causal, yt only depends on x0:t and not on xt+1:T - i.e information from the future is not injected into the past.
Data
- Five‑stock order‑book dataset order-book unknown
Source sentence
Our dataset consisted of five stocks: SPY, GOOG, IBM, VOD, ABEO, trading on five exchanges: NASDAQ, EDGX, BATS, EDGA and ARCA.
Costs
Not applicable
Finds 2
- The weakly supervised model predicts the correct labels with 91% accuracy. predictive-accuracy
Source sentence
We show that we can predict the correct labels with 91% accuracy.
- The proposed framework can rank suspicious order‑book states by similarity to expert‑labelled representations. other
Source sentence
We then conduct a similarity search between any new representation of the order book against the expert labelled representations to rank the results of the weak learner.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 5 related papers). Coverage is partial, so this is neither replication nor refutation.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.
Cited References (6)
- Stock Market Layering Fraud as a Stochastic Optimal Stopping Problem (2019) — link
- MAR (2018) — link
- Searching for Activation Functions (2017) — link
- Release Number 7946-19 (2019) — link
- Microstructure-based manipulation: Strategic behavior and performance of spoofing traders (2013)
- Applied Artificial Intelligence Technology For Processing Trade Data To Detect Patterns Indicative Of Potential Trade Spoofing (2016)
0 of 6 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.