Files
TangCuimin_02141400_2016ofdefense.pdf
UCLouvain restricted access - Adobe PDF
- 4.53 MB
TangCuimin_02141400_2016ofdefense_Annex1.pdf
UCLouvain restricted access - Adobe PDF
- 121.88 KB
Details
- Supervisors
- Faculty
- Degree label
- Abstract
- This thesis aims to analyze the behavior of artificial agents and its impact on the stock price formation based on the multi-agent model. This model is designed to simulate the macro stock market by modeling the micro behavior of each type of agent, which is commonly known as “bottom-up” method. Since the concept of complex adaptive system was introduced from the ecological system into the financial market, the study of the financial market is still a challenge for researchers. For simplicity and tractability, this thesis focuses on studying five types of traders, namely, contrarian traders, momentum traders, herd traders, noise traders and fundamentalists. By simulating their specific trading strategies on Netlogo platform, the stock price will show its trend accordingly. On one hand, the simulated results will be analyzed statistically to test its capacity of explaining these stylized facts in the stock market. On the other hand, agents’ trading behavior will be analyzed by changing some related parameters. Finally, we are able to examine the robustness of the multi-agent simulation and therefore is capability of explaining the macro abnormal phenomena at the micro level.