Adaptive clustering of stock prices data using cascaded competitive learning neural networks

As part of a stock market analysis and prediction system consisting of an expert system and neural networks, clustering of stock prices data is needed. This paper proposes a method of clustering stock prices data using cascaded competitive learning neural networks. Our experiments show that the method has achieved effective clustering results for stock prices data and that the method is easily controlled to produce clustering results which satisfy the customs of stock market analysts. The method can be used in the cases of other data which have intrinsically hierarchical cluster structures.