Multivariate Chaotic Time Series Prediction Based on NARX Neural Networks

DOI

10.2991/eame-17.2017.40How to use a DOI?

Keywords

narx neural networks; chaotic time series; univariate; multivariate; prediction

Abstract

According to multivariate chaotic time series prediction problem, this paper establishes a multivariate chaotic time series forecasting model with nature structure import data based on NARX neural network. The simulation research of Lorenz chaotic time series proves that the forecasting precision of multivariate chaotic time series forecasting model with nature structure import data is much higher than using unvaried chaotic time series. And found the NARX neural network has strong nonlinear mapping ability than others. In comparison with other algorithms, the NARX modeling with nature structure import data method can make better predicting performance, thus it can be widely used in multivariate chaotic time series prediction.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yan Xiu
AU  - Wei Zhang
PY  - 2017/04
DA  - 2017/04
TI  - Multivariate Chaotic Time Series Prediction Based on NARX Neural Networks
BT  - Proceedings of the 2017 2nd International Conference on Electrical, Automation and Mechanical Engineering (EAME 2017)
PB  - Atlantis Press
SP  - 164
EP  - 167
SN  - 2352-5401
UR  - https://doi.org/10.2991/eame-17.2017.40
DO  - 10.2991/eame-17.2017.40
ID  - Xiu2017/04
ER  -