An Improved Platform for Multi-Agent Based Stock Market Simulation in Distributed Environment

Ce YU; Xiang CHEN; Chunyu WANG; Hutong WU; Jizhou SUN; Yuelei LI; Xiaotao ZHANG

doi:10.1587/transinf.2015EDP7050

An Improved Platform for Multi-Agent Based Stock Market Simulation in Distributed Environment

Ce YU, Xiang CHEN, Chunyu WANG, Hutong WU, Jizhou SUN, Yuelei LI, Xiaotao ZHANG

Full Text Views

0

Cite this

Summary :

Multi-agent based simulation has been widely used in behavior finance, and several single-processed simulation platforms with Agent-Based Modeling (ABM) have been proposed. However, traditional simulations of stock markets on single processed computers are limited by the computing capability since financial researchers need larger and larger number of agents and more and more rounds to evolve agents' intelligence and get more efficient data. This paper introduces a distributed multi-agent simulation platform, named PSSPAM, for stock market simulation focusing on large scale of parallel agents, communication system and simulation scheduling. A logical architecture for distributed artificial stock market simulation is proposed, containing four loosely coupled modules: agent module, market module, communication system and user interface. With the customizable trading strategies inside, agents are deployed to multiple computing nodes. Agents exchange messages with each other and with the market based on a customizable network topology through a uniform communication system. With a large number of agent threads, the round scheduling strategy is used during the simulation, and a worker pool is applied in the market module. Financial researchers can design their own financial models and run the simulation through the user interface, without caring about the complexity of parallelization and related problems. Two groups of experiments are conducted, one with internal communication between agents and the other without communication between agents, to verify PSSPAM to be compatible with the data from Euronext-NYSE. And the platform shows fair scalability and performance under different parallelism configurations.

Publication: IEICE TRANSACTIONS on Information Vol.E98-D No.10 pp.1727-1735

Publication Date: 2015/10/01

Publicized: 2015/06/25

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2015EDP7050

Type of Manuscript: PAPER

Category: Fundamentals of Information Systems

Authors

Ce YU
  Tianjin University
Xiang CHEN
  Tianjin University
Chunyu WANG
  Tianjin University
Hutong WU
  Tianjin University
Jizhou SUN
  Tianjin University
Yuelei LI
  Tianjin University
Xiaotao ZHANG
  Tianjin University

Keyword

distributed stock market simulation, agent-based modeling, round scheduling, distributed

Cite this

Copy

Ce YU, Xiang CHEN, Chunyu WANG, Hutong WU, Jizhou SUN, Yuelei LI, Xiaotao ZHANG, "An Improved Platform for Multi-Agent Based Stock Market Simulation in Distributed Environment" in IEICE TRANSACTIONS on Information, vol. E98-D, no. 10, pp. 1727-1735, October 2015, doi: 10.1587/transinf.2015EDP7050.
Abstract: Multi-agent based simulation has been widely used in behavior finance, and several single-processed simulation platforms with Agent-Based Modeling (ABM) have been proposed. However, traditional simulations of stock markets on single processed computers are limited by the computing capability since financial researchers need larger and larger number of agents and more and more rounds to evolve agents' intelligence and get more efficient data. This paper introduces a distributed multi-agent simulation platform, named PSSPAM, for stock market simulation focusing on large scale of parallel agents, communication system and simulation scheduling. A logical architecture for distributed artificial stock market simulation is proposed, containing four loosely coupled modules: agent module, market module, communication system and user interface. With the customizable trading strategies inside, agents are deployed to multiple computing nodes. Agents exchange messages with each other and with the market based on a customizable network topology through a uniform communication system. With a large number of agent threads, the round scheduling strategy is used during the simulation, and a worker pool is applied in the market module. Financial researchers can design their own financial models and run the simulation through the user interface, without caring about the complexity of parallelization and related problems. Two groups of experiments are conducted, one with internal communication between agents and the other without communication between agents, to verify PSSPAM to be compatible with the data from Euronext-NYSE. And the platform shows fair scalability and performance under different parallelism configurations.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2015EDP7050/_p

Copy

@ARTICLE{e98-d_10_1727,
author={Ce YU, Xiang CHEN, Chunyu WANG, Hutong WU, Jizhou SUN, Yuelei LI, Xiaotao ZHANG, },
journal={IEICE TRANSACTIONS on Information},
title={An Improved Platform for Multi-Agent Based Stock Market Simulation in Distributed Environment},
year={2015},
volume={E98-D},
number={10},
pages={1727-1735},
abstract={Multi-agent based simulation has been widely used in behavior finance, and several single-processed simulation platforms with Agent-Based Modeling (ABM) have been proposed. However, traditional simulations of stock markets on single processed computers are limited by the computing capability since financial researchers need larger and larger number of agents and more and more rounds to evolve agents' intelligence and get more efficient data. This paper introduces a distributed multi-agent simulation platform, named PSSPAM, for stock market simulation focusing on large scale of parallel agents, communication system and simulation scheduling. A logical architecture for distributed artificial stock market simulation is proposed, containing four loosely coupled modules: agent module, market module, communication system and user interface. With the customizable trading strategies inside, agents are deployed to multiple computing nodes. Agents exchange messages with each other and with the market based on a customizable network topology through a uniform communication system. With a large number of agent threads, the round scheduling strategy is used during the simulation, and a worker pool is applied in the market module. Financial researchers can design their own financial models and run the simulation through the user interface, without caring about the complexity of parallelization and related problems. Two groups of experiments are conducted, one with internal communication between agents and the other without communication between agents, to verify PSSPAM to be compatible with the data from Euronext-NYSE. And the platform shows fair scalability and performance under different parallelism configurations.},
keywords={},
doi={10.1587/transinf.2015EDP7050},
ISSN={1745-1361},
month={October},}

Copy

TY - JOUR
TI - An Improved Platform for Multi-Agent Based Stock Market Simulation in Distributed Environment
T2 - IEICE TRANSACTIONS on Information
SP - 1727
EP - 1735
AU - Ce YU
AU - Xiang CHEN
AU - Chunyu WANG
AU - Hutong WU
AU - Jizhou SUN
AU - Yuelei LI
AU - Xiaotao ZHANG
PY - 2015
DO - 10.1587/transinf.2015EDP7050
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2015
AB - Multi-agent based simulation has been widely used in behavior finance, and several single-processed simulation platforms with Agent-Based Modeling (ABM) have been proposed. However, traditional simulations of stock markets on single processed computers are limited by the computing capability since financial researchers need larger and larger number of agents and more and more rounds to evolve agents' intelligence and get more efficient data. This paper introduces a distributed multi-agent simulation platform, named PSSPAM, for stock market simulation focusing on large scale of parallel agents, communication system and simulation scheduling. A logical architecture for distributed artificial stock market simulation is proposed, containing four loosely coupled modules: agent module, market module, communication system and user interface. With the customizable trading strategies inside, agents are deployed to multiple computing nodes. Agents exchange messages with each other and with the market based on a customizable network topology through a uniform communication system. With a large number of agent threads, the round scheduling strategy is used during the simulation, and a worker pool is applied in the market module. Financial researchers can design their own financial models and run the simulation through the user interface, without caring about the complexity of parallelization and related problems. Two groups of experiments are conducted, one with internal communication between agents and the other without communication between agents, to verify PSSPAM to be compatible with the data from Euronext-NYSE. And the platform shows fair scalability and performance under different parallelism configurations.
ER -