著者

高橋 大志 寺野 隆雄

出版者

日鉄技術情報センタ-

雑誌

シミュレーション (ISSN:02859947)

巻号頁・発行日

vol.21, no.2, pp.133-142, 2002-06-15

参考文献数

17

被引用文献数

2

---

GARCH (Generalized Autoregressive Conditional Hetero-scedasticity) is a macro level model to estimate the volatility of financial markets. Prospect theory explains micro level characteristics of decision making agents in socio-psychology. Although both concepts are very fundamental in the financial and economic domain and agent-based simulation, however, there are some contradictions among the assumptions of finance theories, agents' internal states, and the behaviors of real markets. In order to bridge the two macro- and micro-level concepts, this paper develops agent-based simulation models. The model consists of simple agents with rational and/or irrational decision making functionalities for investment. The experimental results using both rational and irrational agents have shown that the behaviors of the agents with the characteristics of prospect theory coincide with the estimation by GARCH model. These results have suggested the effectiveness of the agent-based approach to the domain.