Learning the Spoofability of Limit Order Books With Interpretable Probabilistic Neural Networks

2025-04-22

AI Summary

This paper addresses spoofing detection in limit order books (LOBs) of centralized cryptocurrency exchanges, proposing a millisecond-level real-time detection method. Methodologically, it introduces— for the first time—order size and relative distance to the best bid/ask as key features; constructs a multi-scale Hawkes process-driven order flow model; and designs a lightweight probabilistic neural network to directly predict the conditional distribution of mid-price changes, thereby quantifying spoofers’ expected market manipulation gain (spoofability). Contributions include: (1) uncovering the structural impact of order placement distance on price formation; and (2) establishing the first interpretable, probabilistic spoofability framework grounded in manipulation gain—moving beyond conventional volume- or temporal-pattern-based detection paradigms. Evaluated on Level-3 data from December 4–7, 2024, the model identifies 31% of large limit orders as exhibiting statistically significant spoofability, with low deployment overhead suitable for high-frequency trading environments.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Economics, Online Markets and Human Computation: Economic aspects of blockchain and cryptocurrencies
• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity
• Security and Privacy: Large-scale security measurements

Abstract

This paper investigates real-time detection of spoofing activity in limit order books, focusing on cryptocurrency centralized exchanges. We first introduce novel order flow variables based on multi-scale Hawkes processes that account both for the size and placement distance from current best prices of new limit orders. Using a Level-3 data set, we train a neural network model to predict the conditional probability distribution of mid price movements based on these features. Our empirical analysis highlights the critical role of the posting distance of limit orders in the price formation process, showing that spoofing detection models that do not take the posting distance into account are inadequate to describe the data. Next, we propose a spoofing detection framework based on the probabilistic market manipulation gain of a spoofing agent and use the previously trained neural network to compute the expected gain. Running this algorithm on all submitted limit orders in the period 2024-12-04 to 2024-12-07, we find that 31% of large orders could spoof the market. Because of its simple neuronal architecture, our model can be run in real time. This work contributes to enhancing market integrity by providing a robust tool for monitoring and mitigating spoofing in both cryptocurrency exchanges and traditional financial markets.

Problem

• Detect spoofing activity in cryptocurrency limit order books
• Predict mid price movements using neural networks
• Assess market manipulation gain for spoofing detection

Innovation

• Multi-scale Hawkes processes for order flow variables
• Probabilistic neural network for mid price prediction
• Real-time spoofing detection using manipulation gain