Mining better technical trading strategies with genetic algorithms

IEEE Computer Society
Publication Type:
Conference Proceeding
Proceedings of the International Workshop on Integrating AI and Data Mining, 2006, pp. 26 - 33
Issue Date:
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2006005245.pdf1.1 MB
Adobe PDF
Technical analysis is one of the two main schools of thought in the analysis of security prices. It is widely believed and applied by many professional and amateur traders. However, it is often criticized for lacking scientific rigour or worse, for lacking any basis whatsoever. We propose to explore the feasibility and/or limitation of technical analysis by the optimization of technical trading strategies over historical stock data with genetic algorithms. This paper presents the optimization problem in detail and discusses the potential problems to be tackled during the optimization. Preliminary experiments show that it can identify the limitations quickly.
Please use this identifier to cite or link to this item: