Investigating the Size and Value Effect in Determining Performance of Australian Listed Companies: A Neural Network Approach

Luu, J. and Kennedy, P.J.

This paper explores the size and value effect in influencing performance of individual companies using backpropagation neural networks. According to existing theory, companies with small market capitalization and high book to market ratios have a tendency to perform better in the future. Data from over 300 Australian Stock Exchange listed companies between 2000-2004 is examined and a neural network is trained to predict company performance based on market capitalization, book to market ratio, beta and standard deviation. Evidence for the value effect was found over longer time periods but there was less for the size effect. Poor company performance was also observed to be correlated with high risk.

Cite as: Luu, J. and Kennedy, P.J. (2006). Investigating the Size and Value Effect in Determining Performance of Australian Listed Companies: A Neural Network Approach. In Proc. Fifth Australasian Data Mining Conference (AusDM2006), Sydney, Australia. CRPIT, 61. Peter, C., Kennedy, P. J., Li, J., Simoff, S. J. and Williams, G. J., Eds. ACS. 155-161.

(from crpit.com) (local if available)