Investigating neural network modeling decisions for the australian all-ordinaries index

  • Authors:

  • Andrew Flitman;Mark Barnes;Deniss Teng Tai Kiat

  • Affiliations:

  • Monash University, Business Systems, Clayton, Victoria, Australia;Monash University, Computing and IT, Churchill, Victoria, Australia;Monash University, Business Systems, Clayton, Victoria, Australia

  • Venue:
  • ICCS'03 Proceedings of the 2003 international conference on Computational science
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Estimating stock market output depends mainly on identifying non-linear relationships of input variables. To forecast such systems a non-linear modeling tool is required. This paper describes the experimental approaches for developing an Artificial Neural Network for the purpose of modeling the Australian All Ordinaries Index movement over a prediction horizon of 1 year. Network parameters such as network architectures, input data sizing and periodicity are considered in the development of the network. The evaluation criterion for the Neural Network output is the R Square Statistic.