Modeling Financial Markets Based on the Double-threshold Agent Model

Accession number;07A0176808
Title;Modeling Financial Markets Based on the Double-threshold Agent Model
Author; SATO AKIHIRO (Kyoto Univ., JPN)
Journal Title;IPSJ Transactions on Database
Journal Code:Z0778A
ISSN:0387-5806
VOL.48;NO.SIG2(TOM16);PAGE.9-16(2007)
Figure&Table&Reference;FIG.7, TBL.3, REF.20
Pub. Country;Japan
Language;Japanese
Abstract;We analyze tick quotes of the USD/JPY market from 1998 to 2003. Calculating power spectrum densities we find some peaks for them at a few minutes. In order to explain this phenomenon we develop the double-threshold agent model. The double-threshold agent model is a microscopic model of a financial market which consists of agents who determine to buy, sell, or do nothing. Under the assumption that there is a common exogenous periodic information that does not affect decision-making of the agents we find a peak for power spectrum densities of agents' activity at half the frequency of the exogenous information. Furthermore signal-to-noise ratio calculated from them depends on uncertainty of decision-making of agents and has an extreme at optimal strength of the uncertainty, namely, stochastic resonance happens. We propose a hypothesis that appearance and disappearance of the peaks relates to stochastic resonance. (author abst.)