Computational modeling of collective human behavior

The example of financial markets

Andy Kirou, Błazej Ruszczycki, Markus Walser, Neil F Johnson

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

8 Citations (Scopus)

Abstract

As a result of the increased availability of higher precision spatiotemporal datasets, coupled with the realization that most real-world human systems are complex, a new field of computational modeling is emerging in which the goal is to develop minimal models of collective human behavior which are consistent with the observed real-world dynamics in a wide range of systems. For example, in the field of finance, the fluctuations across a wide range of markets are known to exhibit certain generic stylized facts such as a non-Gaussian 'fat-tailed' distribution of price returns. In this paper, we illustrate how such minimal models can be constructed by bridging the gap between two existing, but incomplete, market models: a model in which a population of virtual traders make decisions based on common global information but lack local information from their social network, and a model in which the traders form a dynamically evolving social network but lack any decision-making based on global information. We show that a combination of these two models - in other words, a population of virtual traders with access to both global and local information - produces results for the price return distribution which are closer to the reported stylized facts. Going further, we believe that this type of model can be applied across a wide range of systems in which collective human activity is observed.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages33-41
Number of pages9
Volume5101 LNCS
EditionPART 1
DOIs
StatePublished - 2008
Event8th International Conference on Computational Science, ICCS 2008 - Krakow, Poland
Duration: Jun 23 2008Jun 25 2008

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume5101 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other8th International Conference on Computational Science, ICCS 2008
CountryPoland
CityKrakow
Period6/23/086/25/08

Fingerprint

Collective Behavior
Computational Modeling
Human Behavior
Financial Markets
Social Support
Stylized Facts
Minimal Model
Human Activities
Social Networks
Population
Decision Making
Fats
Range of data
Incomplete Markets
Market Model
Finance
Model
Availability
Fluctuations
Financial markets

Keywords

  • Collective behavior
  • Complex systems
  • Socio-economic systems
  • Virtual traders

ASJC Scopus subject areas

  • Computer Science(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Theoretical Computer Science

Cite this

Kirou, A., Ruszczycki, B., Walser, M., & Johnson, N. F. (2008). Computational modeling of collective human behavior: The example of financial markets. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (PART 1 ed., Vol. 5101 LNCS, pp. 33-41). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5101 LNCS, No. PART 1). https://doi.org/10.1007/978-3-540-69384-0_8

Computational modeling of collective human behavior : The example of financial markets. / Kirou, Andy; Ruszczycki, Błazej; Walser, Markus; Johnson, Neil F.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5101 LNCS PART 1. ed. 2008. p. 33-41 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5101 LNCS, No. PART 1).

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

Kirou, A, Ruszczycki, B, Walser, M & Johnson, NF 2008, Computational modeling of collective human behavior: The example of financial markets. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 1 edn, vol. 5101 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 5101 LNCS, pp. 33-41, 8th International Conference on Computational Science, ICCS 2008, Krakow, Poland, 6/23/08. https://doi.org/10.1007/978-3-540-69384-0_8
Kirou A, Ruszczycki B, Walser M, Johnson NF. Computational modeling of collective human behavior: The example of financial markets. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 1 ed. Vol. 5101 LNCS. 2008. p. 33-41. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). https://doi.org/10.1007/978-3-540-69384-0_8
Kirou, Andy ; Ruszczycki, Błazej ; Walser, Markus ; Johnson, Neil F. / Computational modeling of collective human behavior : The example of financial markets. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5101 LNCS PART 1. ed. 2008. pp. 33-41 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).
@inproceedings{e662b318444c4ffa8d5ff64d60b2b5b4,
title = "Computational modeling of collective human behavior: The example of financial markets",
abstract = "As a result of the increased availability of higher precision spatiotemporal datasets, coupled with the realization that most real-world human systems are complex, a new field of computational modeling is emerging in which the goal is to develop minimal models of collective human behavior which are consistent with the observed real-world dynamics in a wide range of systems. For example, in the field of finance, the fluctuations across a wide range of markets are known to exhibit certain generic stylized facts such as a non-Gaussian 'fat-tailed' distribution of price returns. In this paper, we illustrate how such minimal models can be constructed by bridging the gap between two existing, but incomplete, market models: a model in which a population of virtual traders make decisions based on common global information but lack local information from their social network, and a model in which the traders form a dynamically evolving social network but lack any decision-making based on global information. We show that a combination of these two models - in other words, a population of virtual traders with access to both global and local information - produces results for the price return distribution which are closer to the reported stylized facts. Going further, we believe that this type of model can be applied across a wide range of systems in which collective human activity is observed.",
keywords = "Collective behavior, Complex systems, Socio-economic systems, Virtual traders",
author = "Andy Kirou and Błazej Ruszczycki and Markus Walser and Johnson, {Neil F}",
year = "2008",
doi = "10.1007/978-3-540-69384-0_8",
language = "English (US)",
isbn = "3540693831",
volume = "5101 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 1",
pages = "33--41",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
edition = "PART 1",

}

TY - GEN

T1 - Computational modeling of collective human behavior

T2 - The example of financial markets

AU - Kirou, Andy

AU - Ruszczycki, Błazej

AU - Walser, Markus

AU - Johnson, Neil F

PY - 2008

Y1 - 2008

N2 - As a result of the increased availability of higher precision spatiotemporal datasets, coupled with the realization that most real-world human systems are complex, a new field of computational modeling is emerging in which the goal is to develop minimal models of collective human behavior which are consistent with the observed real-world dynamics in a wide range of systems. For example, in the field of finance, the fluctuations across a wide range of markets are known to exhibit certain generic stylized facts such as a non-Gaussian 'fat-tailed' distribution of price returns. In this paper, we illustrate how such minimal models can be constructed by bridging the gap between two existing, but incomplete, market models: a model in which a population of virtual traders make decisions based on common global information but lack local information from their social network, and a model in which the traders form a dynamically evolving social network but lack any decision-making based on global information. We show that a combination of these two models - in other words, a population of virtual traders with access to both global and local information - produces results for the price return distribution which are closer to the reported stylized facts. Going further, we believe that this type of model can be applied across a wide range of systems in which collective human activity is observed.

AB - As a result of the increased availability of higher precision spatiotemporal datasets, coupled with the realization that most real-world human systems are complex, a new field of computational modeling is emerging in which the goal is to develop minimal models of collective human behavior which are consistent with the observed real-world dynamics in a wide range of systems. For example, in the field of finance, the fluctuations across a wide range of markets are known to exhibit certain generic stylized facts such as a non-Gaussian 'fat-tailed' distribution of price returns. In this paper, we illustrate how such minimal models can be constructed by bridging the gap between two existing, but incomplete, market models: a model in which a population of virtual traders make decisions based on common global information but lack local information from their social network, and a model in which the traders form a dynamically evolving social network but lack any decision-making based on global information. We show that a combination of these two models - in other words, a population of virtual traders with access to both global and local information - produces results for the price return distribution which are closer to the reported stylized facts. Going further, we believe that this type of model can be applied across a wide range of systems in which collective human activity is observed.

KW - Collective behavior

KW - Complex systems

KW - Socio-economic systems

KW - Virtual traders

UR - http://www.scopus.com/inward/record.url?scp=47749131598&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47749131598&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-69384-0_8

DO - 10.1007/978-3-540-69384-0_8

M3 - Conference contribution

SN - 3540693831

SN - 9783540693833

VL - 5101 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 33

EP - 41

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -