Skip to content

Exhaustive Testing of Trader-agents in Realistically Dynamic Continuous Double Auction Markets: AA Does Not Dominate

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Standard

Exhaustive Testing of Trader-agents in Realistically Dynamic Continuous Double Auction Markets : AA Does Not Dominate. / Cliff, Dave.

Proceedings of the 11th International Conference on Autonomous Agents and Artificial Intelligence (ICAART 2019). ed. / Ana Rocha; Luc Steels; Jaap van den Herik. Vol. 2 Prague : SciTePress, 2019. p. 224-236.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Cliff, D 2019, Exhaustive Testing of Trader-agents in Realistically Dynamic Continuous Double Auction Markets: AA Does Not Dominate. in A Rocha, L Steels & J van den Herik (eds), Proceedings of the 11th International Conference on Autonomous Agents and Artificial Intelligence (ICAART 2019). vol. 2, SciTePress, Prague, pp. 224-236, 11th International Conference on Agents and Artificial Intelligence, ICAART 2019, Prague, Czech Republic, 19/02/19. https://doi.org/10.5220/0007382802240236

APA

Cliff, D. (2019). Exhaustive Testing of Trader-agents in Realistically Dynamic Continuous Double Auction Markets: AA Does Not Dominate. In A. Rocha, L. Steels, & J. van den Herik (Eds.), Proceedings of the 11th International Conference on Autonomous Agents and Artificial Intelligence (ICAART 2019) (Vol. 2, pp. 224-236). Prague: SciTePress. https://doi.org/10.5220/0007382802240236

Vancouver

Cliff D. Exhaustive Testing of Trader-agents in Realistically Dynamic Continuous Double Auction Markets: AA Does Not Dominate. In Rocha A, Steels L, van den Herik J, editors, Proceedings of the 11th International Conference on Autonomous Agents and Artificial Intelligence (ICAART 2019). Vol. 2. Prague: SciTePress. 2019. p. 224-236 https://doi.org/10.5220/0007382802240236

Author

Cliff, Dave. / Exhaustive Testing of Trader-agents in Realistically Dynamic Continuous Double Auction Markets : AA Does Not Dominate. Proceedings of the 11th International Conference on Autonomous Agents and Artificial Intelligence (ICAART 2019). editor / Ana Rocha ; Luc Steels ; Jaap van den Herik. Vol. 2 Prague : SciTePress, 2019. pp. 224-236

Bibtex

@inproceedings{60a3599fe31f49fc86ec5cd95af9aba1,
title = "Exhaustive Testing of Trader-agents in Realistically Dynamic Continuous Double Auction Markets: AA Does Not Dominate",
abstract = "We analyse results from over 3.4million detailed market-trading simulation sessions which collectively confirm an unexpected result: in markets with dynamically varying supply and demand, the best-performing automated adaptive auction-market trading-agent currently known in the AI/Agents literature, i.e. Vytelingum’s Adaptive-Aggressive (AA) strategy, can be routinely out-performed by simpler trading strategies. AA is the most recent in a series of AI trading-agent strategies proposed by various researchers over the past twenty years: research papers contributing major steps in this evolution of strategies have been published at IJCAI, in the Artificial Intelligence journal, and at AAMAS. The innovative step taken here is to brute-force exhaustively evaluate AA in market environments that are in various ways more realistic, closer to real-world financial markets, than the simple constrained abstract experimental evaluations routinely used in the prior academic AI/Agents research literature. We conclude that AA can indeed appear dominant when tested only against other AI-based trading agents in the highly simplified market scenarios that have become the methodological norm in the trading-agents academic research literature, but much of that success seems to be because AA was designed with exactly those simplified experimental markets in mind. As soon as we put AA in scenarios closer to real-world markets, modify it to fit those markets accordingly, and exhaustively test it against simpler trading agents, AA’s dominance simply disappears.",
keywords = "Agent-based Computational Economics, Automated Trading, Computational Finance, Financial Markets",
author = "Dave Cliff",
year = "2019",
month = "3",
day = "14",
doi = "10.5220/0007382802240236",
language = "English",
isbn = "9789897583506",
volume = "2",
pages = "224--236",
editor = "Ana Rocha and Luc Steels and {van den Herik}, Jaap",
booktitle = "Proceedings of the 11th International Conference on Autonomous Agents and Artificial Intelligence (ICAART 2019)",
publisher = "SciTePress",

}

RIS - suitable for import to EndNote

TY - GEN

T1 - Exhaustive Testing of Trader-agents in Realistically Dynamic Continuous Double Auction Markets

T2 - AA Does Not Dominate

AU - Cliff, Dave

PY - 2019/3/14

Y1 - 2019/3/14

N2 - We analyse results from over 3.4million detailed market-trading simulation sessions which collectively confirm an unexpected result: in markets with dynamically varying supply and demand, the best-performing automated adaptive auction-market trading-agent currently known in the AI/Agents literature, i.e. Vytelingum’s Adaptive-Aggressive (AA) strategy, can be routinely out-performed by simpler trading strategies. AA is the most recent in a series of AI trading-agent strategies proposed by various researchers over the past twenty years: research papers contributing major steps in this evolution of strategies have been published at IJCAI, in the Artificial Intelligence journal, and at AAMAS. The innovative step taken here is to brute-force exhaustively evaluate AA in market environments that are in various ways more realistic, closer to real-world financial markets, than the simple constrained abstract experimental evaluations routinely used in the prior academic AI/Agents research literature. We conclude that AA can indeed appear dominant when tested only against other AI-based trading agents in the highly simplified market scenarios that have become the methodological norm in the trading-agents academic research literature, but much of that success seems to be because AA was designed with exactly those simplified experimental markets in mind. As soon as we put AA in scenarios closer to real-world markets, modify it to fit those markets accordingly, and exhaustively test it against simpler trading agents, AA’s dominance simply disappears.

AB - We analyse results from over 3.4million detailed market-trading simulation sessions which collectively confirm an unexpected result: in markets with dynamically varying supply and demand, the best-performing automated adaptive auction-market trading-agent currently known in the AI/Agents literature, i.e. Vytelingum’s Adaptive-Aggressive (AA) strategy, can be routinely out-performed by simpler trading strategies. AA is the most recent in a series of AI trading-agent strategies proposed by various researchers over the past twenty years: research papers contributing major steps in this evolution of strategies have been published at IJCAI, in the Artificial Intelligence journal, and at AAMAS. The innovative step taken here is to brute-force exhaustively evaluate AA in market environments that are in various ways more realistic, closer to real-world financial markets, than the simple constrained abstract experimental evaluations routinely used in the prior academic AI/Agents research literature. We conclude that AA can indeed appear dominant when tested only against other AI-based trading agents in the highly simplified market scenarios that have become the methodological norm in the trading-agents academic research literature, but much of that success seems to be because AA was designed with exactly those simplified experimental markets in mind. As soon as we put AA in scenarios closer to real-world markets, modify it to fit those markets accordingly, and exhaustively test it against simpler trading agents, AA’s dominance simply disappears.

KW - Agent-based Computational Economics

KW - Automated Trading

KW - Computational Finance

KW - Financial Markets

UR - http://www.scopus.com/inward/record.url?scp=85064807989&partnerID=8YFLogxK

U2 - 10.5220/0007382802240236

DO - 10.5220/0007382802240236

M3 - Conference contribution

AN - SCOPUS:85064807989

SN - 9789897583506

VL - 2

SP - 224

EP - 236

BT - Proceedings of the 11th International Conference on Autonomous Agents and Artificial Intelligence (ICAART 2019)

A2 - Rocha, Ana

A2 - Steels, Luc

A2 - van den Herik, Jaap

PB - SciTePress

CY - Prague

ER -