The investigation of the agent in the artificial market

Takahiro Kitakubo, Yusuke Koyama, Hiroshi Deguchi

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

Abstract

In this paper, we investigate the investment strategy in the artificial market called U-Mart, which is designed to provide a common test bed for researchers in the fields of economics and information sciences. UMIE is the international experiment of U-Mart as a contests of trading agents. We attended UMIE 2003 and 2004, and our agent won the championship in both experiments. We examin why this agent is strong in UMIE environment. The strategy of this agent is called "on-line learning" or "real-time learning". Concretely, the agent exploits and forecasts futures price fluctuations by means of identifying the environment in reinforcement learning. We examined an efficiency of price forecasting in the classified environment. To examine the efficacy of it, we executed experiments 1000 times with UMIE open-type simulation standard toolkits, and we verified that forecasting futures price fluctuation in our strategy is useful for better trading.

Original languageEnglish
Title of host publicationLecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)
EditorsT.G. Kim
Pages215-223
Number of pages9
Volume3397
Publication statusPublished - 2005
Externally publishedYes
Event13th International Conference on AIS 2004 - Jeju Island, Korea, Republic of
Duration: 2004 Oct 42004 Oct 6

Other

Other13th International Conference on AIS 2004
CountryKorea, Republic of
CityJeju Island
Period04/10/404/10/6

Fingerprint

Information science
Experiments
Reinforcement learning
Economics

Keywords

  • Artificial Market
  • Reinforcement Learning
  • U-Mart

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

Kitakubo, T., Koyama, Y., & Deguchi, H. (2005). The investigation of the agent in the artificial market. In T. G. Kim (Ed.), Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 3397, pp. 215-223)

The investigation of the agent in the artificial market. / Kitakubo, Takahiro; Koyama, Yusuke; Deguchi, Hiroshi.

Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). ed. / T.G. Kim. Vol. 3397 2005. p. 215-223.

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

Kitakubo, T, Koyama, Y & Deguchi, H 2005, The investigation of the agent in the artificial market. in TG Kim (ed.), Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). vol. 3397, pp. 215-223, 13th International Conference on AIS 2004, Jeju Island, Korea, Republic of, 04/10/4.
Kitakubo T, Koyama Y, Deguchi H. The investigation of the agent in the artificial market. In Kim TG, editor, Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). Vol. 3397. 2005. p. 215-223
Kitakubo, Takahiro ; Koyama, Yusuke ; Deguchi, Hiroshi. / The investigation of the agent in the artificial market. Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). editor / T.G. Kim. Vol. 3397 2005. pp. 215-223
@inproceedings{a8d8cb2a368549b1b2477265887c0190,
title = "The investigation of the agent in the artificial market",
abstract = "In this paper, we investigate the investment strategy in the artificial market called U-Mart, which is designed to provide a common test bed for researchers in the fields of economics and information sciences. UMIE is the international experiment of U-Mart as a contests of trading agents. We attended UMIE 2003 and 2004, and our agent won the championship in both experiments. We examin why this agent is strong in UMIE environment. The strategy of this agent is called {"}on-line learning{"} or {"}real-time learning{"}. Concretely, the agent exploits and forecasts futures price fluctuations by means of identifying the environment in reinforcement learning. We examined an efficiency of price forecasting in the classified environment. To examine the efficacy of it, we executed experiments 1000 times with UMIE open-type simulation standard toolkits, and we verified that forecasting futures price fluctuation in our strategy is useful for better trading.",
keywords = "Artificial Market, Reinforcement Learning, U-Mart",
author = "Takahiro Kitakubo and Yusuke Koyama and Hiroshi Deguchi",
year = "2005",
language = "English",
volume = "3397",
pages = "215--223",
editor = "T.G. Kim",
booktitle = "Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)",

}

TY - GEN

T1 - The investigation of the agent in the artificial market

AU - Kitakubo, Takahiro

AU - Koyama, Yusuke

AU - Deguchi, Hiroshi

PY - 2005

Y1 - 2005

N2 - In this paper, we investigate the investment strategy in the artificial market called U-Mart, which is designed to provide a common test bed for researchers in the fields of economics and information sciences. UMIE is the international experiment of U-Mart as a contests of trading agents. We attended UMIE 2003 and 2004, and our agent won the championship in both experiments. We examin why this agent is strong in UMIE environment. The strategy of this agent is called "on-line learning" or "real-time learning". Concretely, the agent exploits and forecasts futures price fluctuations by means of identifying the environment in reinforcement learning. We examined an efficiency of price forecasting in the classified environment. To examine the efficacy of it, we executed experiments 1000 times with UMIE open-type simulation standard toolkits, and we verified that forecasting futures price fluctuation in our strategy is useful for better trading.

AB - In this paper, we investigate the investment strategy in the artificial market called U-Mart, which is designed to provide a common test bed for researchers in the fields of economics and information sciences. UMIE is the international experiment of U-Mart as a contests of trading agents. We attended UMIE 2003 and 2004, and our agent won the championship in both experiments. We examin why this agent is strong in UMIE environment. The strategy of this agent is called "on-line learning" or "real-time learning". Concretely, the agent exploits and forecasts futures price fluctuations by means of identifying the environment in reinforcement learning. We examined an efficiency of price forecasting in the classified environment. To examine the efficacy of it, we executed experiments 1000 times with UMIE open-type simulation standard toolkits, and we verified that forecasting futures price fluctuation in our strategy is useful for better trading.

KW - Artificial Market

KW - Reinforcement Learning

KW - U-Mart

UR - http://www.scopus.com/inward/record.url?scp=26844490194&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=26844490194&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:26844490194

VL - 3397

SP - 215

EP - 223

BT - Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)

A2 - Kim, T.G.

ER -