Adaptive polynomial neural networks for times series forecasting

Panos Liatsis, Amalia Foka, John Yannis Goulermas, Lidija Mandic

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Scopus citations

Abstract

Time series prediction involves the determination of an appropriate model, which can encapsulate the dynamics of the system, described by the sample data. Previous work has demonstrated the potential of neural networks in predicting the behaviour of complex, non-linear systems. In particular, the class of polynomial neural networks has been shown to possess universal approximation properties, while ensuring robustness to noise and missing data, good generalisation and rapid learning. In this work, a polynomial neural network is proposed, whose structure and weight values are determined with the use of evolutionary computing. The resulting networks allow an insight into the relationships underlying the input data, hence allowing a qualitative analysis of the models' performance. The approach is tested on a variety of non-linear time series data.

Original languageBritish English
Title of host publicationProceedings ELMAR-2007 - 49th International Symposium ELMAR-2007 focused on Mobile Multimedia
Pages35-39
Number of pages5
DOIs
StatePublished - 2007
Event49th International Symposium ELMAR-2007 focused on Mobile Multimedia - Zadar, Croatia
Duration: 12 Sep 200714 Sep 2007

Publication series

NameProceedings Elmar - International Symposium Electronics in Marine
ISSN (Print)1334-2630

Conference

Conference49th International Symposium ELMAR-2007 focused on Mobile Multimedia
Country/TerritoryCroatia
CityZadar
Period12/09/0714/09/07

Keywords

  • Forecasting
  • Genetic algorithms
  • Polynomial neural networks
  • Time series

Fingerprint

Dive into the research topics of 'Adaptive polynomial neural networks for times series forecasting'. Together they form a unique fingerprint.

Cite this