Time series prediction based on NARX neural networks: An advanced approach

The NARX network is a dynamical neural architecture commonly used for input-output modeling of nonlinear dynamical systems. When applied to time series prediction, the NARX network is designed as a feedforward time delay neural network (TDNN), i.e., without the feedback loop of delayed outputs, redu...

Full description

Saved in:
Bibliographic Details
Published in2009 International Conference on Machine Learning and Cybernetics Vol. 3; pp. 1275 - 1279
Main Authors Hang Xie, Hao Tang, Yu-He Liao
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.07.2009
Subjects
Artificial neural networks
Computer architecture
Cybernetics
Delay effects
Embedding theory
Gamma Test
Machine learning
NARX networks
Neural networks
Neurons
Predictive models
Testing
Time series prediction
Vibration measurement
Online AccessGet full text

Cover

More Information
Summary:The NARX network is a dynamical neural architecture commonly used for input-output modeling of nonlinear dynamical systems. When applied to time series prediction, the NARX network is designed as a feedforward time delay neural network (TDNN), i.e., without the feedback loop of delayed outputs, reducing substantially its predictive performance. In this paper, it is shown that the original architecture of the NARX network can be easily and efficiently applied to prediction of time series using embedding theory to reconstruct the input of NARX network. We evaluate the proposed approach using a real-world data set, which is the vibration data measured from a Co2 compressor. The results show that the proposed approach consistently outperforms standard neural network based predictors, such as the TDNN architecture.
ISBN:9781424437023
1424437024
ISSN:2160-133X
DOI:10.1109/ICMLC.2009.5212326