Competitive advantage for multiple-memory strategies in an artificial market

Kurt E. Mitman, Sehyo Charley Choe, Neil F Johnson

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

Abstract

We consider a simple binary market model containing N competitive agents. The novel feature of our model is that it incorporates the tendency shown by traders to look for patterns in past price movements over multiple time scales, i.e. multiple memory-lengths. In the regime where these memory-lengths are all small, the average winnings per agent exceed those obtained for either (1) a pure population where all agents have equal memory-length, or (2) a mixed population comprising sub-populations of equal-memory agents with each sub-population having a different memory-length. Agents who consistently play strategies of a given memory-length, are found to win more on average - switching between strategies with different memory lengths incurs an effective penalty, while switching between strategies of equal memory does not. Agents employing short-memory strategies can outperform agents using long-memory strategies, even in the regime where an equal-memory system would have favored the use of long-memory strategies. Using the many-body 'Crowd-Anticrowd' theory, we obtain analytic expressions which are in good agreement with the observed numerical results. In the context of financial markets, our results suggest that multiple-memory agents have a better chance of identifying price patterns of unknown length and hence will typically have higher winnings.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsD. Abbott, J.-P. Bouchaud, X. Gabaix, J.L. McCauley
Pages225-232
Number of pages8
Volume5848
DOIs
StatePublished - 2005
Externally publishedYes
EventNoise and Fluctuations in Econophysics and Finance - Austin, TX, United States
Duration: May 24 2005May 26 2005

Other

OtherNoise and Fluctuations in Econophysics and Finance
CountryUnited States
CityAustin, TX
Period5/24/055/26/05

Fingerprint

Data storage equipment
penalties
tendencies
Computer systems

Keywords

  • Econophysics
  • Limited resources
  • Multi-agent games
  • Prediction

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Mitman, K. E., Choe, S. C., & Johnson, N. F. (2005). Competitive advantage for multiple-memory strategies in an artificial market. In D. Abbott, J-P. Bouchaud, X. Gabaix, & J. L. McCauley (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 5848, pp. 225-232). [30] https://doi.org/10.1117/12.618869

Competitive advantage for multiple-memory strategies in an artificial market. / Mitman, Kurt E.; Choe, Sehyo Charley; Johnson, Neil F.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / D. Abbott; J.-P. Bouchaud; X. Gabaix; J.L. McCauley. Vol. 5848 2005. p. 225-232 30.

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

Mitman, KE, Choe, SC & Johnson, NF 2005, Competitive advantage for multiple-memory strategies in an artificial market. in D Abbott, J-P Bouchaud, X Gabaix & JL McCauley (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 5848, 30, pp. 225-232, Noise and Fluctuations in Econophysics and Finance, Austin, TX, United States, 5/24/05. https://doi.org/10.1117/12.618869
Mitman KE, Choe SC, Johnson NF. Competitive advantage for multiple-memory strategies in an artificial market. In Abbott D, Bouchaud J-P, Gabaix X, McCauley JL, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 5848. 2005. p. 225-232. 30 https://doi.org/10.1117/12.618869
Mitman, Kurt E. ; Choe, Sehyo Charley ; Johnson, Neil F. / Competitive advantage for multiple-memory strategies in an artificial market. Proceedings of SPIE - The International Society for Optical Engineering. editor / D. Abbott ; J.-P. Bouchaud ; X. Gabaix ; J.L. McCauley. Vol. 5848 2005. pp. 225-232
@inproceedings{55925a2710b34983b8fd18cc0dec0b6c,
title = "Competitive advantage for multiple-memory strategies in an artificial market",
abstract = "We consider a simple binary market model containing N competitive agents. The novel feature of our model is that it incorporates the tendency shown by traders to look for patterns in past price movements over multiple time scales, i.e. multiple memory-lengths. In the regime where these memory-lengths are all small, the average winnings per agent exceed those obtained for either (1) a pure population where all agents have equal memory-length, or (2) a mixed population comprising sub-populations of equal-memory agents with each sub-population having a different memory-length. Agents who consistently play strategies of a given memory-length, are found to win more on average - switching between strategies with different memory lengths incurs an effective penalty, while switching between strategies of equal memory does not. Agents employing short-memory strategies can outperform agents using long-memory strategies, even in the regime where an equal-memory system would have favored the use of long-memory strategies. Using the many-body 'Crowd-Anticrowd' theory, we obtain analytic expressions which are in good agreement with the observed numerical results. In the context of financial markets, our results suggest that multiple-memory agents have a better chance of identifying price patterns of unknown length and hence will typically have higher winnings.",
keywords = "Econophysics, Limited resources, Multi-agent games, Prediction",
author = "Mitman, {Kurt E.} and Choe, {Sehyo Charley} and Johnson, {Neil F}",
year = "2005",
doi = "10.1117/12.618869",
language = "English (US)",
volume = "5848",
pages = "225--232",
editor = "D. Abbott and J.-P. Bouchaud and X. Gabaix and J.L. McCauley",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Competitive advantage for multiple-memory strategies in an artificial market

AU - Mitman, Kurt E.

AU - Choe, Sehyo Charley

AU - Johnson, Neil F

PY - 2005

Y1 - 2005

N2 - We consider a simple binary market model containing N competitive agents. The novel feature of our model is that it incorporates the tendency shown by traders to look for patterns in past price movements over multiple time scales, i.e. multiple memory-lengths. In the regime where these memory-lengths are all small, the average winnings per agent exceed those obtained for either (1) a pure population where all agents have equal memory-length, or (2) a mixed population comprising sub-populations of equal-memory agents with each sub-population having a different memory-length. Agents who consistently play strategies of a given memory-length, are found to win more on average - switching between strategies with different memory lengths incurs an effective penalty, while switching between strategies of equal memory does not. Agents employing short-memory strategies can outperform agents using long-memory strategies, even in the regime where an equal-memory system would have favored the use of long-memory strategies. Using the many-body 'Crowd-Anticrowd' theory, we obtain analytic expressions which are in good agreement with the observed numerical results. In the context of financial markets, our results suggest that multiple-memory agents have a better chance of identifying price patterns of unknown length and hence will typically have higher winnings.

AB - We consider a simple binary market model containing N competitive agents. The novel feature of our model is that it incorporates the tendency shown by traders to look for patterns in past price movements over multiple time scales, i.e. multiple memory-lengths. In the regime where these memory-lengths are all small, the average winnings per agent exceed those obtained for either (1) a pure population where all agents have equal memory-length, or (2) a mixed population comprising sub-populations of equal-memory agents with each sub-population having a different memory-length. Agents who consistently play strategies of a given memory-length, are found to win more on average - switching between strategies with different memory lengths incurs an effective penalty, while switching between strategies of equal memory does not. Agents employing short-memory strategies can outperform agents using long-memory strategies, even in the regime where an equal-memory system would have favored the use of long-memory strategies. Using the many-body 'Crowd-Anticrowd' theory, we obtain analytic expressions which are in good agreement with the observed numerical results. In the context of financial markets, our results suggest that multiple-memory agents have a better chance of identifying price patterns of unknown length and hence will typically have higher winnings.

KW - Econophysics

KW - Limited resources

KW - Multi-agent games

KW - Prediction

UR - http://www.scopus.com/inward/record.url?scp=28544436448&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=28544436448&partnerID=8YFLogxK

U2 - 10.1117/12.618869

DO - 10.1117/12.618869

M3 - Conference contribution

VL - 5848

SP - 225

EP - 232

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Abbott, D.

A2 - Bouchaud, J.-P.

A2 - Gabaix, X.

A2 - McCauley, J.L.

ER -