High-Frequency Trading (HFT), from its inception in the early 2000s to its current usage, is a set of technologies that enables high-speed trading by specialized investment firms, which can execute multiple trades within milliseconds (Pandey and Wu, 2015). This opens a new range of possibilities, such as the ability to perform arbitrage, inflate prices, anticipate market movements, among others (Breckenfelder, 2024).
However, when technological and regulatory forces equalize, these firms are exposed to classic dilemmas from game theory, which, if unresolved, can lead to millions in losses (Jones, 2013).
A basic example of HFT usage is index arbitrage. The S&P 500 is an index generated on the Chicago Mercantile Exchange, while $SPY is a ticket that replicates the S&P 500.
However, the replication is no automatic, and the ticket appears after the index. If at any point there is a delay in the replication, it could be exploited as a signal of a price change in $SPY. Due to the high speeds of HFT firms, they can take advantage of a spread, generating perhaps only a few cents in profit, but at high volume (Baldauf and Mollner, 2020).
Taking account these characteristics about HFT firms, we can infer that there is a vast field of action for the application of Game Theory. For example, the following diagram shows a flow for implementing an A/B test with a starting point based on Game Theory.
Triveni et al., (2022) make a comparison between the performance of different types of machine learning models vs. Game Theory analysis, which shows a clear advantage for the latter.
Why did Game Theory receive higher recognition? One reason could be that Game Theory is a rational method for resolving conflicts based on the interaction of features, in this case involving high-speed interactions.
---------------
Author: David Tello is a second-year student in the MSc in Economic Engineering program at the National University of Engineering. His research interests include financial markets and trading.
References
[1] Markus Baldauf and Joshua Mollner. High-frequency trading and market performance. The Journal of Finance, 75(3):1495–1526, 2020.
[2] Johannes Breckenfelder. Competition among high-frequency traders and market quality. Journal of Eco- nomic Dynamics and Control, 166:104922, 2024.
[3] Charles M Jones. What do we know about high-frequency trading? Columbia Business School Research Paper, (13-11), 2013.
[4] Vivek K Pandey and Chen Y Wu. Investors may take heart: A game theoretic view of high frequency trading. Journal of Financial Planning, 28(5), 2015.
[5] VS Triveni, T Deepthi, and MP Molimol. Optimal game theory model for stock price prediction. Mathematical Statistician and Engineering Applications, 71(4):3043–3054, 2022.
Comments