A Machine Learning Attack on Illegal Trading

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.