Open Access Repository

Application of Higher-Order Neural Networks to Financial Time-Series Prediction


Downloads per month over past year

Fulcher, J, Zhang, M and Xu, S 2006 , 'Application of Higher-Order Neural Networks to Financial Time-Series Prediction', in Joarder Kamruzzaman and Rezaul Begg and Ruhul Sarker (eds.), Artificial Neural Networks in Finance and Manufacturing , Idea Group Publishing, Hershey, PA.

kamruzzaman_Xu_...pdf | Download (1MB)
Available under University of Tasmania Standard License.

| Preview


Financial time series data is characterized by non-linearities, discontinuities and high frequency, multi-polynomial components. Not surprisingly, conventional Artificial Neural Networks (ANNs) have difficulty in modelling such complex data. A more appropriate approach is to apply Higher-Order ANNs, which are capable of extracting higher order polynomial coefficients in the data. Moreover, since there is a one-to-one correspondence between network weights and polynomial coefficients, HONNs (unlike ANNs generally) can be considered open-, rather than 'closed box' solutions, and thus hold more appeal to the financial community. After developing Polynomial and Trigonometric HONNs, we introduce the concept of HONN groups. The latter incorporate piecewise continuous activation functions and thresholds, and as a result are capable of modelling discontinuous (piecewise continuous) data, and what's more to any degree of accuracy. Several other PHONN variants are also described. The performance of P(T)HONNs and HONN groups on representative financial time series is described (credit ratings and exchange rates). In short, HONNs offer roughly twice the performance of MLP/BP on financial time series prediction, and HONN groups around 10% further improvement.

Item Type: Book Section
Authors/Creators:Fulcher, J and Zhang, M and Xu, S
Keywords: artificial neural networks, financial time series, higher-order artificial neural networks, artificial neural network group
Publisher: Idea Group Publishing
Item Statistics: View statistics for this item

Actions (login required)

Item Control Page Item Control Page