Forecasting with the aid of hybrid models, combining neural networks and nonparametric regression models in time series

Mammadov M., Aydin D.

24th Mini EURO Conference on Continuous Optimization and Information-Based Technologies in the Financial Sector, MEC EurOPT 2010, İzmir, Turkey, 23 - 26 June 2010, pp.281-287 identifier

  • Publication Type: Conference Paper / Full Text
  • City: İzmir
  • Country: Turkey
  • Page Numbers: pp.281-287
  • Keywords: Hybrid models, Multilayer perceptrons, Radial basis function, Regression spline, Smoothing spline
  • Anadolu University Affiliated: Yes


The goal of this article is to introduce the hybrid models that combine 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. In the paper, we fully developed a new hybrid model where the first component is smoothing spline and the second component is multilayer perceptron. The performance of this new model is compared by forecasting two real Turkish data sets: Domestic product per capita (GDP) and the number of tourist arrivals. The results obtained by experimental evaluations show that hybrid model developed in this paper have performed much better in comparison to hybrid models discussed in the literature by others. © Izmir University of Economics, Turkey, 2010.