Predicting the stock market to maximise returns for the small investor : a data mining approach

Abstract

Investors in the stock market are always interested and looking for better methods of predicting the direction of stock price movements with a view to reduce the risk of investment and to improve on the capital gains. This research introduces two algorithms to achieve this objective and evaluate their power of prediction. These algorithms are then generalised to develop a new data mining methodology in the area of "pattern recognition" to identify extreme events in a time varying and real time data stream analysis. Further a software tool is specifically developed as part of this research to perform the constructed algorithms and analyse and evaluate the effectiveness of these methods in predicting stock market movements.

Date of Award	2009
Original language	English