A Machine Learning Attack on Illegal Trading

James, Robert; Leung, Henry; Prokhorov, Artem

doi:10.2139/ssrn.3722391

A Machine Learning Attack on Illegal Trading

58 Pages Posted: 21 Jan 2021 Last revised: 25 Jan 2021

See all articles by Robert James

Henry Leung

University of Sydney Business School; Financial Research Network (FIRN)

Artem Prokhorov

University of Sydney Business School; Saint Petersburg State University; University of Montreal - Centre interuniversitaire de recherche en économie quantitative (CIREQ)

Date Written: October 31, 2020

Abstract

We design an adaptive framework for the detection of illegal trading behavior. Its key component is an extension of a pattern recognition tool, originating from the field of signal processing and adapted to modern electronic systems of securities trading. The new methodology combines the flexibility of dynamic time warping with contemporary approaches from extreme value theory to explore large-scale order book data and accurately identify illegal trading patterns without access to any confirmed illegal transactions for training. The method is shown to achieve remarkable improvements over alternative approaches in the identification of suspected illegal insider trading cases in a high-frequency dataset provided by an international investment firm.

Keywords: Market Manipulation, Dynamic Time Warping, Market Microstructure, Machine Learning, Limit Order Books

JEL Classification: C02, C4, G14

Suggested Citation: Suggested Citation

James, Robert and Leung, Henry and Prokhorov, Artem, A Machine Learning Attack on Illegal Trading (October 31, 2020). Available at SSRN: https://ssrn.com/abstract=3722391 or http://dx.doi.org/10.2139/ssrn.3722391