Market abuse detection is the use of surveillance systems and analytics to identify illegal trading behaviours such as insider dealing, spoofing, layering, and price manipulation. It analyses order-book activity, trade patterns, and communications to flag anomalies for compliance review and regulatory reporting. It is a core obligation under regimes such as the EU Market Abuse Regulation and is increasingly powered by machine learning.
Content
- Detection engines combine rule-based scenarios for known typologies with unsupervised anomaly detection and network analysis to surface novel abuse patterns. False-positive reduction is a central challenge, driving adoption of supervised models trained on confirmed cases and explainable scoring for investigator triage.