Hybrid Models Combining Neural Networks and Nonparametric Regression Models Used for Time Series Prediction

AYDIN D., Mammadov M.

9th WSEAS International Conference on Systems Theory and Scientific Computation, Moscow, Russia, 20 - 22 August 2009, pp.141-143, (Full Text) identifier identifier

  • Publication Type: Conference Paper / Full Text
  • City: Moscow
  • Country: Russia
  • Page Numbers: pp.141-143
  • Keywords: Time series, Neural networks, Multilayer perceptrons, Radial basis function, Nonparametric regression, Smoothing spline, Regression spline, Additive regression model, Hybrid models, ARIMA
  • Anadolu University Affiliated: Yes

Abstract

In this paper, we proposed the hybrid models whose components are nonparametric regression and artificial neural networks. Smoothing spline, regression spline and additive regression models are considered as the nonparametric regression components. Furthermore, various multilayer perceptron algorithms and radial basis function network model are regarded as the artificial neural networks components. The performances of the models have been compared for the number of cars produced in Turkey. The results obtained by experimental evaluations show that hybrid models proposed in this study have performed much better in comparison to hybrid models examined by others (see for example, [1] and [2]).