Date of this Version


Document Type

Conference Paper

Publication Details

This paper was presented at AISAT 2004 – International Conference on Artificial Intelligence in Science and Technology, held at the School of Engineering, The University of Tasmania, Hobart, Australia, 21 to 24 November 2004.


This paper demonstrates how Neural Networks may be successfully applied to the problem of security selection in the Australian stockmarket. In practice, it is unrealistic for a trader to apply capital to all securities available in the market, and a selection technique must be employed to reduce the number of securities competing for capital. Selection techniques are generally based on either fundamental analysis procedures, or technical analysis procedures. This paper focuses on fundamental procedures, and implements a neural network which enhances the effectiveness of these procedures.



This document has been peer reviewed.