Advertisement

Applied Intelligence

, Volume 37, Issue 4, pp 511–519 | Cite as

An enhanced hybrid method for time series prediction using linear and neural network models

  • Purwanto
  • C. Eswaran
  • R. Logeswaran
Article

Abstract

The need for improving the accuracy of time series prediction has motivated researchers to develop more efficient prediction models. The accuracy rates resulting from linear models such as linear regression (LR), exponential smoothing (ES) and autoregressive integrated moving average (ARIMA) are not high as they are poor in handling the nonlinear time series data. Neural network models are considered to be better in handling such nonlinear time series data. In the real-world problems, the time series data consist of complex linear and nonlinear patterns and it may be difficult to obtain high prediction accuracy rates using only linear or neural network models. Hybrid models which combine both linear and neural network models can be used to obtain high prediction accuracy rates. In this paper, we propose an enhanced hybrid model which indicates for a given input data which choice is better between the two options, namely, a linear-nonlinear combination or a nonlinear-linear combination. The appropriate combination is selected based on a linearity test of data. From the experimental results, it is found that the proposed hybrid model comprising linear-nonlinear combination performs better than other models for the data that have a linear relationship. On the contrary, the hybrid model comprising nonlinear-linear combination performs better than other models for the data that have a nonlinear relationship.

Keywords

Exponential smoothing Linear regression ARIMA Neural network Enhanced hybrid method 

References

  1. 1.
    Javier C, Rosario E, Francisco JN, Conejo AJ (2003) ARIMA models to predict next-day electricity prices. IEEE Trans Power Syst 18:1014–1020 CrossRefGoogle Scholar
  2. 2.
    Katimon A, Demun AS (2004) Water use trend at universiti teknologi Malaysia: application of ARIMA model. J Pendidik Univ Teknol Malays 41:47–56 Google Scholar
  3. 3.
    Moghaddas-Tafreshi SM, Farhadi M (2008) Linear regression-based study for temperature sensitivity analysis of Iran electrical load. In: IEEE international conference on industrial technology, pp 1–7 Google Scholar
  4. 4.
    Li BJ, Hua HC (2007) The combined forecasting method of GM(1,1) with linear regression and its application. In: International conference on grey systems and intelligent services, pp 394–398 Google Scholar
  5. 5.
    Sitte R (2002) Neural networks approach to the random walk dilemma of financial time series. Appl Intell 16:163–171 zbMATHCrossRefGoogle Scholar
  6. 6.
    Neukukar V, Hamidi-Beheshti MT (2010) A local linear radial basis function neural network for financial time series forecasting. Appl Intell 33:352–356 CrossRefGoogle Scholar
  7. 7.
    Harri N, Teri H, Ari K, Juhani R, Kolehmaine M (2004) Evolving the neural network model for forecasting air pollution time series. Eng Appl Artif Intell 17:159–167 CrossRefGoogle Scholar
  8. 8.
    Faruk DO (2010) A hybrid neural network and ARIMA model for water quality time series prediction. Eng Appl Artif Intell 23:586–594 CrossRefGoogle Scholar
  9. 9.
    Santoso S (2009) Business forecasting: metode peramalan bisnis masa kini dengan minitab dan SPSS. Elex Media Komputindo, Jakarta Google Scholar
  10. 10.
    Le CT (2003) Introductory biostatistics. Wiley, New York zbMATHCrossRefGoogle Scholar
  11. 11.
    Peter JB, Davis RA (2002) Introduction to time series and forecasting. Springer, New York zbMATHGoogle Scholar
  12. 12.
    Suhartono (2008) Feedforward neural networks untuk pemodelan runtun waktu. Dissertation, Gajah Mada University, Yogyakarta, Indonesia Google Scholar
  13. 13.
    Denton JW (1995) How good are neural networks for causal forecasting? J Bus Forecast 14:17–20 Google Scholar
  14. 14.
    Zhang GP (2003) Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing 50:159–175 zbMATHCrossRefGoogle Scholar
  15. 15.
    Roselina S, Siti MS, Hashim SZM (2008) Hybridization model of linear and nonlinear time series data for forecasting. In: Second Asia international conference on modelling & simulation, pp 597–602 Google Scholar
  16. 16.
    Kim TH, Lee YS, Newbold P (2004) Spurious nonlinear regressions in econometrics. School of Economics, University of Nottingham Google Scholar
  17. 17.
    World Health Organization (2009) Global tuberculosis control: epidemiology, strategy, financing: WHO report 2009. WHO Press, Geneva Google Scholar
  18. 18.
    Hyndman RJ (n.d.) (2011) Time series data library. http://robjhyndman.com/TSDL. Accessed on October 10, 2011
  19. 19.
    Rojas I, Valenzuela O, Rojas F, Guillen A, Herrera LJ, Pomares H et al (2008) Soft-computing techniques and ARMA model for time series prediction. Neurocomputing 71:519–537 CrossRefGoogle Scholar
  20. 20.
    Lee YS, Tong LI (2011) Forecasting time series using a methodology based on autoregressive integrated moving average and genetic programming. Knowl-Based Syst 24:66–72 CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2012

Authors and Affiliations

  1. 1.Faculty of Information TechnologyMultimedia UniversityCyberjayaMalaysia
  2. 2.Faculty of EngineeringMultimedia UniversityCyberjayaMalaysia
  3. 3.Faculty of Computer ScienceDian Nuswantoro UniversitySemarangIndonesia

Personalised recommendations