Nonlinear time series forecasting with dynamic RBF neural networks

Author

Zhang, Dongqing ; Ning, Xuanxi ; Liu, Xueni ; Han, Yubing

Author_Institution

Coll. of Econ.&Manage., Nanjing Univ. of Aeronaut. & Astronaut., Nanjing

fYear

2008

fDate

25-27 June 2008

Firstpage

6988

Lastpage

6993

Abstract

In order to cope with nonlinear time series, a variable structure radial basis function (RBF) networks model, in which the numbers of basis functions and input order vary over time, is proposed in this paper. Then sequential Monte Carlo (SMC) method is used for time series on-line prediction and corresponding algorithm is developed. At last, the data of weekly price of the shipbuilding steel product are analyzed, and experimental results indicate that the variable structure RBF networks model proposed is effective.

Keywords

Monte Carlo methods; forecasting theory; radial basis function networks; time series; RBF neural networks; SMC; nonlinear time series forecasting; radial basis function; sequential Monte Carlo method; time series online prediction; Autoregressive processes; Economic forecasting; Intelligent control; Monte Carlo methods; Neural networks; Prediction algorithms; Predictive models; Radial basis function networks; Sliding mode control; Steel; Prediction; Radial Basis Function Neural Networks; Sequential Monte Carlo Methods;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation, 2008. WCICA 2008. 7th World Congress on

Conference_Location

Chongqing

Print_ISBN

978-1-4244-2113-8

Electronic_ISBN

978-1-4244-2114-5

Type

conf

DOI

10.1109/WCICA.2008.4593999

Filename

4593999