Loading...
Neuro-fuzzy model for stock market prediction
Author(s)
Date Issued
December 1, 1999
Type
Conference Paper
Abstract
Stock value prediction is a widely concerned issue. In the stock market, every investor certainly would like to be able to precisely forecast stock prices in order to maximize their profit and reduce potential risks. The tools used by stock market analysts, such as moving average and trend techniques, can only give investors an alarm or a sign of possible increase or decrease in prices. These traditional mathematical approaches are not very effective in forecasting the future stock value because they are not susceptible to a change in circumstances, especially for Thai stock market which is very small in volume and easily disturbed by the outside environments, e.g. political issues and tumors. In this study, the new neuro-fuzzy architecture is proposed. The proposed system is employed to build the model to predict the future values of the Krung Thai Bank PLC, Thailand's largest government-owned bank. The results demonstrate a very reliable performance of the proposed model.
Citation
Intelligent Engineering Systems Through Artificial Neural Networks, 9, 587-591, 1999