Using Learning Classifier System for Making Investment Strategies based on Institutional Analysis

J.-Y. Lin, C.-P. Cheng, W.-C. Tsai, and A.-P. Chen (Taiwan)


learning classifier system, institutional investment, stockmarket


The artificial intelligence can dynamically learn and adapt to the change of environments for maximizing the desired goals. This paper conducts simulation experiment to evolve learning classifier system (LCS) for short-term stock forecast decision. Since stock price trend is influenced by unknown and unpredictable surroundings, using LCS to model the fluctuations on financial market allows for the capability to discover the patterns of future trends. Furthermore, the institutional investment is the main consideration of this research by implementing LCS for making strategies. More specifically, LCS is capable of evolving from generation to generation, and in this way can provide the highest profit for future decision-making. In simulation work using real financial data, it is found that LCS produces great profits, and is quite practical for investors.

Important Links:

Go Back