Incorporation of statistical methods in multi-step neural network prediction models

Author

Cloarec, Guy-Michel ; Ringwood, John

Author_Institution

Sch. of Electron. Eng., Dublin City Univ., Ireland

Volume

3

fYear

1998

fDate

4-9 May 1998

Firstpage

2513

Abstract

This paper addresses the problems associated with multistep ahead prediction neural networks models. We will see how some concepts from the statistical theory field can be applied in various ways to improve the modelling. The generalization and error autocorrelation problems will he addressed using topological and methodological approach among which network committees, statistical bootstrap and principal component analysis will play a key role. These methods will be applied to the sunspot time series

Keywords

correlation theory; forecasting theory; generalisation (artificial intelligence); neural nets; statistical analysis; PCA; error autocorrelation; generalization; multistep-ahead prediction neural networks models; network committees; principal component analysis; statistical bootstrap; statistical methods; statistical theory; sunspot time series; topology; Autocorrelation; Helium; Intelligent networks; Network topology; Neural networks; Optimization methods; Predictive models; Principal component analysis; Statistical analysis; Sun;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on

Conference_Location

Anchorage, AK

ISSN

1098-7576

Print_ISBN

0-7803-4859-1

Type

conf

DOI

10.1109/IJCNN.1998.687257

Filename

687257