Nonlinear time series forecasting with Bayesian neural networks

Kocadaǧli O., Aşikgil B.

Expert Systems with Applications, vol.41, no.15, pp.6596-6610, 2014 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 41 Issue: 15
  • Publication Date: 2014
  • Doi Number: 10.1016/j.eswa.2014.04.035
  • Journal Name: Expert Systems with Applications
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.6596-6610
  • Keywords: Bayesian neural networks, Gaussian approximation, Genetic algorithms, Hybrid Monte Carlo simulations, Nonlinear time series, Recursive hyperparameters
  • Istanbul Kültür University Affiliated: No

Abstract

The Bayesian learning provides a natural way to model the nonlinear structure as the artificial neural networks due to their capability to cope with the model complexity. In this paper, an evolutionary Monte Carlo (MC) algorithm is proposed to train the Bayesian neural networks (BNNs) for the time series forecasting. This approach called as Genetic MC is based on Gaussian approximation with recursive hyperparameter. Genetic MC integrates MC simulations with the genetic algorithms and the fuzzy membership functions. In the implementations, Genetic MC is compared with the traditional neural networks and time series techniques in terms of their forecasting performances over the weekly sales of a Finance Magazine. © 2014 Elsevier Ltd. All rights reserved.