Refinement of generated weighted fuzzy production rules by using fuzzy neural networks for stock market prediction.

Abstract

One of the most important problems in the modern finance is finding efficient ways of summarizing the stock market data that would allow one to obtain useful information about the behavior of the market. The trader's expectations to predict stock markets are seriously affected by some uncertain factors including political situation, oil price, overall world situation, local stock markets etc. Therefore, predicting stock price movements is quite difficult. In this paper, the new technique to predict stock market is presented for the refinement of generated Weighted Fuzzy Production Rules (WFPR's) by using fuzzy neural networks. The existing techniques to generate WFPR's are suffered from the problem of low accuracy of classifying or recognizing unseen examples. The reasons for having these problems are 1) the WFPRs generated are not powerful enough to represent the domain knowledge, 2) the techniques used to generate WFPRs are pre- matured, ad-hoc or may not be suitable for the prediction problem, and 3) further refinement of the extracted rules has not been done. In this paper, we look into the solutions of the above problems by 1) enhancing the representation power of WFPRs by including local and global weights, 2) developing a fuzzy neural network (FNN) with enhanced learning algorithm, and 3) using this FNN to refine the local and global weights of WFPRs for stock market prediction. By experiment our method with some stock markets examples has found a better accuracy in classifying unseen samples without increasing the number of extracted WFPRs.