Presentation 2004/3/8
Conditions for Genetic Algorithm Learning Describes Investor Sentiment
Takashi YAMADA, Kazuhiro UEDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The purpose of this paper is to clarify whether the Genetic Algorithm Learning can describe the Model of Investor Sentiment(Barberis et al., J. of Financial Economics, 49, pp.307-343, 1998), one of the studies of the Behavioral Finance. For this purpose, we explored the conditions using the agents' viewpoints towards market which were obtained when some series of typical asset-returns were given. As a result, some conditions for genetic algorithm were shown to be required: First, in order to represent the model of investor sentiment by genetic algorithm learning, agents need to know market condition for their learning. Second, the information used when agents select their parents must be up-to-date.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) agent-based computational economics / genetic algorithm / learning / investor sentiment
Paper # AI2003-80
Date of Issue

Conference Information
Committee AI
Conference Date 2004/3/8(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Vice Chair

Paper Information
Registration To Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Conditions for Genetic Algorithm Learning Describes Investor Sentiment
Sub Title (in English)
Keyword(1) agent-based computational economics
Keyword(2) genetic algorithm
Keyword(3) learning
Keyword(4) investor sentiment
1st Author's Name Takashi YAMADA
1st Author's Affiliation Department of General Systems Studies, Graduate School of Arts and Sciences, University of Tokyo()
2nd Author's Name Kazuhiro UEDA
2nd Author's Affiliation Interfaculty Initiative of Information Studies, University of Tokyo
Date 2004/3/8
Paper # AI2003-80
Volume (vol) vol.103
Number (no) 724
Page pp.pp.-
#Pages 6
Date of Issue