Simulating the Investor’s Behaviors in Stock Market with an Intelligent Agent

• Main Authors: Feng-Yu Chang, 張鳳玉
• Other Authors: Deng-Yiv Chiu
• Format: Others
• Language: en_US
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/12448050188047005638

碩士 === 中華大學 === 電機工程學系碩士班 === 90 === In this project, from the investor’s perspective, we will use intelligent agents to simulate the investor’s behaviors. The intelligent agent will learn the investor’s behavior according to investor’s historical data. It will learn from the decisions making for the investment. The factors that influence investors are complicated. These factors include external environments, individual wealth, profits of the company, messages scattered in the investment markets and so on. Also different investors would have different views about these messages, and investor’s anticipated mentalities are different so that they may make different decisions. In this paper, we propose a framework using an intelligent agent to simulate the individual investor’s behaviors. And we will build a virtual stock market in which intelligent agents can invest according to their decisions. Intelligent agents’ investment decisions are different due to different consideration and conclusion for the same message. Then, every intelligent agent will deal with these messages. We will train intelligent agent with data mining technology and artificial neural network to make the intelligent agent’s behaviors similar to investor’s behaviors in the real world. The intelligent agent makes decisions and invests in the virtual stock market.