Polynomial Pipelined Neural Network and Its Application to Financial Time Series Prediction

A novel type of higher order pipelined neural network, the polynomial pipelined neural network, is presented. The network is constructed from a number of higher order neural networks concatenated with each other to predict highly nonlinear and nonstationary signals based on the engineering concept of divide and conquer. It is evaluated in financial time series application to predict the exchange rate between the US Dollar and 3 other currencies. The network demonstrates more accurate forecasting and an improvement in the signal to noise ratio over a number of benchmarked neural network.

• Type: Conference
• Year: 2006
• Where: AUSAI
• Authors: Abir Jaafar Hussain, Adam Knowles, Paulo J. G. Lisboa, Wael El-Deredy, Dhiya Al-Jumeily
• Keywords: Artificial Intelligence, AUSAI 2006, Financial Time Series, Neural Network, Order Neural Networks