Advertisement

Agent Collaboration for Multiple Trading Strategy Integration

  • Longbing Cao
  • Dan Luo
  • Yanshan Xiao
  • Zhigang Zheng
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4953)

Abstract

The collaboration of agents can undertake complicated tasks that cannot be handled well by a single agent. This is even true for excecuting multiple goals at the same time. In this paper, we demonstrate the use of trading agent collaboration in integrating multiple trading strategies. Trading agents are used for developing quality trading strategies to support smart actions in the market. Evolutionary trading agents are armed with evolutionary computing capability to optimize strategy parameters. To develop even smarter trading strategies (we call golden strategies), multiple Evolutionary and Collaborative trading agents negotiate with each other for m loops to search multiple local strategies with best parameter combinations. They also integrate multiple classes of strategies for trading agents to achieve the best global objectives acceptable for trader needs. Tests of five classes of trading strategies in ten years of five markets of data have shown that agent collaboration for strategy integration can achieve much better performance of trading compared with that of either individually optimized or randomly chosen strategies.

Keywords

Multiagent System Trading Strategy Strategy Agent Coordinator Agent Candidate Strategy 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
    Trading Agent Competition: http://www.sics.se/tac/
  6. 6.
    Trading Agent Design and Analysis: http://www.sics.se/tada05/
  7. 7.
    Cao, L.B.: Multi-strategy integration for actionable trading agents. In: Proceedings of the International Conference on Intelligent Agent Technology, IEEE Computer Society Press, Los Alamitos (2007)Google Scholar
  8. 8.
    Cao, L.B., Zhang, C.Q.: Two-Way Significance of Knowledge Actionability. Int. J. of Business Intelligence and Data Mining (2007)Google Scholar
  9. 9.
    Cao, L.B., Zhang, C.Q.: F-Trade: An Agent-Mining Symbiont for Financial Services. In: Proceedings of the International Conference on Autonomous Agents and Multiagent Systems, ACM Press, New York (2007)Google Scholar
  10. 10.
    Cao, L.B., Luo, C., Zhang, C.Q.: Developing Actionable Trading Strategies for Trading Agents. In: Proceedings of the International Conference on Intelligent Agent Technology (2007)Google Scholar
  11. 11.
    Cao, L.B., Luo, D., Zhang, C.Q.: Fuzzy Genetic Algorithms for Pairs Mining. In: Yang, Q., Webb, G. (eds.) PRICAI 2006. LNCS (LNAI), vol. 4099, pp. 711–720. Springer, Heidelberg (2006)Google Scholar
  12. 12.
    Chan, T.: Artificial Markets and Intelligent Agents, PhD thesis, Massachusetts Institute of Technology (2001)Google Scholar
  13. 13.
    Wan, H.A., Hunter, A.: On Artificial Adaptive Agents Models of Stock Markets. Simulation 68, 279–289 (1997)Google Scholar
  14. 14.
    Wellman, M.P., et al.: Designing the Market Game for a Trading Agent Competition. IEEE Internet Computing 5(2), 43–51 (2001)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Longbing Cao
    • 1
  • Dan Luo
    • 1
  • Yanshan Xiao
    • 2
  • Zhigang Zheng
    • 3
  1. 1.Faculty of Information TechnologyUniversity of TechnologySydneyAustralia
  2. 2.Department of Information TechnologyGuangzhou Asian Games Organizing CommitteeChina
  3. 3.Department of Information TechnologyPICC, Property And Casualty LtdChina

Personalised recommendations