Machine learning, knowledge risk, and principal-agent problems in automated trading

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Machine learning, knowledge risk, and principal-agent problems in automated trading. / Borch, Christian.

In: Technology in Society, Vol. 68, 01.02.2022, p. 101852.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Borch, C 2022, 'Machine learning, knowledge risk, and principal-agent problems in automated trading', Technology in Society, vol. 68, pp. 101852. https://doi.org/10.1016/j.techsoc.2021.101852

APA

Borch, C. (2022). Machine learning, knowledge risk, and principal-agent problems in automated trading. Technology in Society, 68, 101852. https://doi.org/10.1016/j.techsoc.2021.101852

Vancouver

Borch C. Machine learning, knowledge risk, and principal-agent problems in automated trading. Technology in Society. 2022 Feb 1;68:101852. https://doi.org/10.1016/j.techsoc.2021.101852

Author

Borch, Christian. / Machine learning, knowledge risk, and principal-agent problems in automated trading. In: Technology in Society. 2022 ; Vol. 68. pp. 101852.

Bibtex

@article{e0760a5b9a8545f8b6705e023bfe04fe,
title = "Machine learning, knowledge risk, and principal-agent problems in automated trading",
abstract = "Present-day securities trading is dominated by fully automated algorithms. These algorithmic systems are characterized by particular forms of knowledge risk (adverse effects relating to the use or absence of certain forms of knowledge) and principal-agent problems (goal conflicts and information asymmetries arising from the delegation of decision-making authority). Where automated trading systems used to be based on human-defined rules, increasingly, machine-learning (ML) techniques are being adopted to produce machine-generated strategies. Drawing on 213 interviews with market participants involved in automated trading, this study compares the forms of knowledge risk and principal-agent relations characterizing both human-defined and ML-based automated trading systems. It demonstrates that certain forms of ML-based automated trading lead to a change in knowledge risks, particularly concerning dramatically changing market settings, and that they are characterized by a lack of insight into how and why trading rules are being produced by the ML systems. This not only intensifies but also reconfigures principal-agent problems in financial markets.",
keywords = "Faculty of Social Sciences, Automated trading, Financial markets, Knowledge risk, Machine learning, Principal-agent problems",
author = "Christian Borch",
year = "2022",
month = feb,
day = "1",
doi = "10.1016/j.techsoc.2021.101852",
language = "English",
volume = "68",
pages = "101852",
journal = "Technology in Society",
issn = "0160-791X",
publisher = "Elsevier Limited",

}

RIS

TY - JOUR

T1 - Machine learning, knowledge risk, and principal-agent problems in automated trading

AU - Borch, Christian

PY - 2022/2/1

Y1 - 2022/2/1

N2 - Present-day securities trading is dominated by fully automated algorithms. These algorithmic systems are characterized by particular forms of knowledge risk (adverse effects relating to the use or absence of certain forms of knowledge) and principal-agent problems (goal conflicts and information asymmetries arising from the delegation of decision-making authority). Where automated trading systems used to be based on human-defined rules, increasingly, machine-learning (ML) techniques are being adopted to produce machine-generated strategies. Drawing on 213 interviews with market participants involved in automated trading, this study compares the forms of knowledge risk and principal-agent relations characterizing both human-defined and ML-based automated trading systems. It demonstrates that certain forms of ML-based automated trading lead to a change in knowledge risks, particularly concerning dramatically changing market settings, and that they are characterized by a lack of insight into how and why trading rules are being produced by the ML systems. This not only intensifies but also reconfigures principal-agent problems in financial markets.

AB - Present-day securities trading is dominated by fully automated algorithms. These algorithmic systems are characterized by particular forms of knowledge risk (adverse effects relating to the use or absence of certain forms of knowledge) and principal-agent problems (goal conflicts and information asymmetries arising from the delegation of decision-making authority). Where automated trading systems used to be based on human-defined rules, increasingly, machine-learning (ML) techniques are being adopted to produce machine-generated strategies. Drawing on 213 interviews with market participants involved in automated trading, this study compares the forms of knowledge risk and principal-agent relations characterizing both human-defined and ML-based automated trading systems. It demonstrates that certain forms of ML-based automated trading lead to a change in knowledge risks, particularly concerning dramatically changing market settings, and that they are characterized by a lack of insight into how and why trading rules are being produced by the ML systems. This not only intensifies but also reconfigures principal-agent problems in financial markets.

KW - Faculty of Social Sciences

KW - Automated trading

KW - Financial markets

KW - Knowledge risk

KW - Machine learning

KW - Principal-agent problems

U2 - 10.1016/j.techsoc.2021.101852

DO - 10.1016/j.techsoc.2021.101852

M3 - Journal article

VL - 68

SP - 101852

JO - Technology in Society

JF - Technology in Society

SN - 0160-791X

ER -

ID: 319888591