A Study of Artificial Financial Market
Abstract
Initially, financial market research has focused on analytical frameworks that are based on the assumption of rational behavior. However, recent studies have pointed out that actual markets do not seem to show rational behavior. Therefore, in the last decade there has been an increasing interest in describing financial markets using agent-based models. By investigating different market structures, such as trading mechanisms and agent characteristics, this approach provides a variety of insights about why the efficient market hypothesis does not always hold true. This paper also follows an agent-based modeling approach to build a simple financial market. The objective of this study is to build a simple market mechanism and then analyze how traders with different initial trading biases shift their biases as stock prices and strategies change during the simulation. Theoretical models that are frequently used in computational finance and market micro structure are used to construct the simulation model.
Recommended Citation
N. H. Kilicay et al., "A Study of Artificial Financial Market," Intelligent Systems through Artificial Neural Networks Smart Engineering Systems Design; Neural Networks, Evolutionary Programming and Artificial Life, American Society of Mechanical Engineers (ASME), Jan 2005.
Department(s)
Engineering Management and Systems Engineering
Keywords and Phrases
Agent-Based Financial Market Architecture; Agent-Based Modeling; Computational Finance; Market Micro Structure; Multi-Agent Modeling
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2005 American Society of Mechanical Engineers (ASME), All rights reserved.
Publication Date
01 Jan 2005