skip to main content
Explore Scholarly Publications and Datasets in the NSF-PAR

Title: Market Manipulation: An Adversarial Learning Framework for Detection and Evasion

We propose an adversarial learning framework to capture the evolving game between a regulator who develops tools to detect market manipulation and a manipulator who obfuscates actions to evade detection. The model includes three main parts: (1) a generator that learns to adapt original manipulation order streams to resemble trading patterns of a normal trader while preserving the manipulation intent; (2) a discriminator that differentiates the adversarially adapted manipulation order streams from normal trading activities; and (3) an agent-based simulator that evaluates the manipulation effect of adapted outputs. We conduct experiments on simulated order streams associated with a manipulator and a market-making agent respectively. We show examples of adapted manipulation order streams that mimic a specified market maker's quoting patterns and appear qualitatively different from the original manipulation strategy we implemented in the simulator. These results demonstrate the possibility of automatically generating a diverse set of (unseen) manipulation strategies that can facilitate the training of more robust detection algorithms.
Authors:
;
Award ID(s):
1741190
Publication Date:
NSF-PAR ID:
10185932
Journal Name:
29th International Joint Conference on Artificial Intelligence
Page Range or eLocation-ID:
4626 to 4632
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ( , Games)
    We present an agent-based model of manipulating prices in financial markets through spoofing: submitting spurious orders to mislead traders who learn from the order book. Our model captures a complex market environment for a single security, whose common value is given by a dynamic fundamental time series. Agents trade through a limit-order book, based on their private values and noisy observations of the fundamental. We consider background agents following two types of trading strategies: the non-spoofable zero intelligence (ZI) that ignores the order book and the manipulable heuristic belief learning (HBL) that exploits the order book to predict price outcomes.more »We conduct empirical game-theoretic analysis upon simulated agent payoffs across parametrically different environments and measure the effect of spoofing on market performance in approximate strategic equilibria. We demonstrate that HBL traders can benefit price discovery and social welfare, but their existence in equilibrium renders a market vulnerable to manipulation: simple spoofing strategies can effectively mislead traders, distort prices and reduce total surplus. Based on this model, we propose to mitigate spoofing from two aspects: (1) mechanism design to disincentivize manipulation; and (2) trading strategy variations to improve the robustness of learning from market information. We evaluate the proposed approaches, taking into account potential strategic responses of agents, and characterize the conditions under which these approaches may deter manipulation and benefit market welfare. Our model provides a way to quantify the effect of spoofing on trading behavior and market efficiency, and thus it can help to evaluate the effectiveness of various market designs and trading strategies in mitigating an important form of market manipulation.« less
  2. ; ; ( , ACM International Conference on AI in Finance)
    We study learning-based trading strategies in markets where prices can be manipulated through spoofing: the practice of submitting spurious orders to mislead traders who use market information. To reduce the vulnerability of learning traders to such manipulation, we propose two variations based on the standard heuristic belief learning (HBL) trading strategy, which learns transaction probabilities from market activities observed in an order book. The first variation selectively ignores orders at certain price levels, particularly where spoof orders are likely to be placed. The second considers the full order book, but adjusts its limit order price to correct for bias inmore »decisions based on the learned heuristic beliefs. We employ agent-based simulation to evaluate these variations on two criteria: effectiveness in non-manipulated markets and robustness against manipulation. Background traders can adopt (non-learning) zero intelligence strategies or HBL, in its basic form or the two variations. We conduct empirical game-theoretic analysis upon simulated payoffs to derive approximate strategic equilibria, and compare equilibrium outcomes across a variety of trading environments. Results show that agents can strategically make use of the option to block orders to improve robustness against spoofing, while retaining a comparable competitiveness in non-manipulated markets. Our second HBL variation exhibits a general improvement over standard HBL, in markets with and without manipulation. Further explorations suggest that traders can enjoy both improved profitability and robustness by combining the two proposed variations.« less
  3. ; ; ( , ACM International Conference on AI in Finance)
    We study learning-based trading strategies in markets where prices can be manipulated through spoofing: the practice of submitting spurious orders to mislead traders who use market information. To reduce the vulnerability of learning traders to such manipulation, we propose two variations based on the standard heuristic belief learning (HBL) trading strategy, which learns transaction probabilities from market activities observed in an order book. The first variation selectively ignores orders at certain price levels, particularly where spoof orders are likely to be placed. The second considers the full order book, but adjusts its limit order price to correct for bias inmore »decisions based on the learned heuristic beliefs. We employ agent-based simulation to evaluate these variations on two criteria: effectiveness in non-manipulated markets and robustness against manipulation. Background traders can adopt the (non-learning) zero intelligence strategies or HBL, in its basic form or the two variations. We conduct empirical game-theoretic analysis upon simulated payoffs to derive approximate strategic equilibria, and compare equilibrium outcomes across a variety of trading environments. Results show that agents can strategically make use of the option to block orders to improve robustness against spoofing, while retaining a comparable competitiveness in non-manipulated markets. Our second HBL variation exhibits a general improvement over standard HBL, in markets with and without manipulation. Further explorations suggest that traders can enjoy both improved profitability and robustness by combining the two proposed variations.« less
  4. ; ; ( , SIGSIM-PADS '20: Proceedings of the 2020 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation)
    We introduce ABIDES, an open source Agent-Based Interactive Discrete Event Simulation environment. ABIDES is designed from the ground up to support agent-based research in market applications. While proprietary simulations are available within trading firms, there are no broadly available high-fidelity market simulation environments. ABIDES enables the simulation of tens of thousands of trading agents interacting with an exchange agent to facilitate transactions. It supports configurable pairwise noisy network latency between each individual agent as well as the exchange. Our simulator's message-based design is modeled after NASDAQ's published equity trading protocols ITCH and OUCH. We introduce the design of the simulatormore »and illustrate its use and configuration with sample code, validating the environment with example trading scenarios. The utility of ABIDES for financial research is illustrated through experiments to develop a market impact model. The core of ABIDES is a general-purpose discrete event simulation, and we demonstrate its breadth of application with a non-finance work-in-progress simulating secure multiparty federated learning. We close with discussion of additional experimental problems it can be, or is being, used to explore, such as the development of machine learning trading algorithms. We hope that the availability of such a platform will facilitate research in this important area.« less
  5. ; ; ; ( , Proceedings of the 2021 ASEE Annual Conference & Exposition)
    This design-focused practice paper presents a case study describing how a training program developed for academic contexts was adapted for use with engineers working in industry. The underlying curriculum is from the NSF-funded CyberAmbassadors program, which developed training in communication, teamwork and leadership skills for participants from academic and research settings. For the case study described here, one module from the CyberAmbassadors project was adapted for engineers working in private industry: “Teaming Up: Effective Group and Meeting Management.” The key objectives were to increase knowledge and practical skills within the company’s engineering organization, focusing specifically on time management as itmore »relates to project and product delivery. We were also interested in examining the results of translating curricula designed for an academic setting into a corporate setting. Training participants were all from the dedicated engineering department of a US-based location of an international company that provides financial services. The original curriculum was designed for live, in-person training, but was adapted for virtual delivery after the company adopted a 100% remote workforce in response to the COVID-19 pandemic. The training was conducted in four phases: (1) train-the-trainer to create internal evangelists; (2) train management to build buy-in and provide sponsorship; (3) phased rollout of training to individual members of the engineering department, contemporaneous with (4) specific and intentional opportunities to apply the skills in normal business activities including Joint Architecture Design (JAD) sessions. Effectiveness was measured through surveys at the engineering management level (before, during, and after training), and through direct discussions with engineering teams who were tracked for four weeks after the training. A number of cultural shifts within the company were observed as direct and indirect outcomes of this training. These include the creation and standardization of a template for meeting agendas; a “grassroots” effort to spread the knowledge and best practices from trained individuals to untrained individuals through informal, peer-to-peer interactions; individuals at varying levels of company hierarchy publicly expressing that they would not attend meetings unless an appropriate agenda was provided in advance; and requests for additional training by management who wanted to increase performance in their employees. As a result of this adaptation from academic to industry training contexts, several key curricular innovations were added back to the original CyberAmbassadors corpus. Examples include a reinterpretation of the separate-but-equal leadership roles within meetings, and the elevation of timekeeper to a controlling leadership role within a meeting. This case study offers valuable lessons on translating training from academic/research settings to industry, including a description of how the “business case” was developed in order to gain approval for the training and sponsorship from management. Future work includes adapting additional material from the CyberAmbassadors program for applications in a business context, and the continued formal and informal propagation of the current material within the company.« less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email