Market manipulation in a high-frequency context: colliding particle physics tools and financial market data

Summary

The digitalization of financial markets creates new possibilities in trading, such as the automation of actions and algorithmic trading. This has led to new types of market manipulation that occur faster than human eyes can see. Academics and regulators experience difficulties in applying current analytical techniques to identify market manipulation, partly due to the vast amounts of data produced by these markets. In addition, the legal framework provides insufficient guidelines to distinguish between legitimate and illegal behavior. These challenges present a distinct issue that this dissertation addresses: analyzing and identifying market manipulation in a high-frequency context. We reap the benefits from the decades of big data experience from the European Organization for Nuclear Research (CERN) by applying particle physics tools to financial market data. The focus is on the market manipulation practice of ‘spoofing’ in U.S. futures markets. The dissertation demonstrates 1) that the particle physics data-analysis framework ROOT is effective for storing, processing, visualizing and analyzing high-frequency data; 2) strategies to delineate, characterize, identify, visualize and analyze market manipulation; and 3) the synthesis of these approaches. Thereby, providing tools for academics and regulators to accelerate market surveillance developments.