The authors propose a method of construction of transaction strategies in capital markets, which utilizes a Kohonen neural network called also self-organizing map (SOM). This network is applied for cluster analysis of input vectors that describe particular states of a stock market. The presented approach is quite different than typical neural techniques used for financial markets modeling or prediction, as it utilizes a network trained with an unsupervised method. The paper submits also the research concerning the effectiveness of the described method, based on analysis of active investment strategy for stocks of KGHM (Polish mining and metallurgy corporation).
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.