NON-STATIONARY TIME SERIES FORECASTING BASED ON MULTIWAVELET POLYMORPHIC NETWORK

There are many methods and models for forecasting non-stationary time series. However, the problem of the accuracy and adequacy of the forecast of non-stationary time series has not been solved yet. In this paper, a new forecast model, based on a multiwavelet network with additional customizable parameters, which is called polymorphic, is proposed. The efficiency of the proposed model is compared with the well-known time series forecast models like autoregressive integrated moving average model, multilayer perceptron and hybrid model in which both models are combined. Three well-known real data sets (the Wolf's sunspot data, the Canadian lynx data and the British pound/US dollar exchange rate data) were taken as empirical data. The comparison showed that forecast model based on the proposed multiwavelet polymorphic network has a smaller prediction error for each series. This is achieved by introducing additional customizable parameters into the wavelet network, which allow to better adapt to the non-stationary nature of time series. Moreover, for the wavelet network to perform well in the presence of linearity, were used linear connections between the wavelet neurons of input and output layers. The proposed technology can be used to predict the time series generated by dynamic processes of a different nature.

1. Box, G., Jenkins, G. M., Reinsel, G.C., Ljung, G.M. Time Series Analysis: Forecasting and Control. Hoboken, New Jersey, 2015. 712 p.

2. Box, G., Jenkins, G. M., Reinsel, G.C., Ljung, G.M. Time Series Analysis: Forecasting and Control. Hoboken, New Jersey, 2015. 712 p.

3. Ratnadip A., Agrawal R.K. Homogeneous Ensemble of Artificial Neural Networks for Time Series Forecasting // International Journal of Computer Applications. 2011 vol. 32 no. 7. pp. 1-8.

4. Cortez P., Rio M., Rocha M., Sousa P. Multi-scale Internet traffic forecasting using neural networks and time series methods // Expert Systems. 2012. vol. 29. no.2. 143-155

5. Benkachcha S., Benhra J., Hassani H. Seasonal Time Series Forecasting Models based on Artificial Neural Network // International Journal of Computer Applications. 2015. vol. 75 no. 7 pp. 37-42.

6. Zhang G. Time Series Forecasting Using a Hybrid ARIMA and Neural Network Model // Neurocomputing. 2003. vol. 50, pp. 159-175.

7. Zhang G.P. A neural network ensemble method with jittered training data for time series forecasting // Information Sciences. 2007. vol. 177. no. 23. pp. 5329-5346

8. Khashei M., Bijari M. An artificial neural network (p, d,q) model for timeseries forecasting // Expert Systems with Applications. 2010. vol. 37. no. 1. pp. 479–489.

9. Ina K., Ratnadip A., Ghanshyam V. Time Series Forecasting Using Hybrid ARIMA and ANN Models Based on DWT Decomposition // Procedia Computer Science. 2015. vol.48.pp. 173-179.

10. Daubechies I. The wavelet transform, time-frequency localization and signal analysis // IEEE Transactions on Information Theory. 1990. vol. 36.no.5.pp.961–1005.

11. Lokenath D., Firdous A.S., “Wavelet Transforms and Their Applications”, Springer New York Heidelberg Dordrecht London, 304 p., 2015.

12. Kriechbaumer T., Angus A., Parsons D., Casado M.R., 2014. An improved wavelet–ARIMA approach for forecasting metal prices // Resour. Policy. 2014.vol. 39.no.1. pp. 32–41.

13. Fard A.K., Zadeh M.R.A. A hybrid method based on wavelet, ANN and ARIMA model for short-term load forecasting // J. Exp. Theor. Artif. Intell. 2014.vol.26.no.2. pp.167–182.

14. Ramana, R.V., Krishna, B., Kumar, S.R., Pandey, N.G. Monthly rainfall prediction using wavelet neural network analysis // Water Resour. Manage. 2013.vol. 27.no.10. pp.3697–3711.

15. Alexandridis A.K., Zapranis A.D. Wavelet neural networks: A practical guide // Neural Networks. 2013. vol. 42. pp.1-27

16. Zhao J., Chen B., Shen J. Multidimensional non-orthogonal wavelet-sigmoid basis function neural network for dynamic process fault diagnosis // Computers and Chemical Engineering. 1998.vol.23.no.7.pp.83–92.

17. Verzunov S.N. Synthesis of the polymorphic wavelet network and investigation of its properties for approximation of time-dependent time series [Informatika i sistemy upravlenija].2015.no. 2.pp.60–69. (In Russian)

18. Verzunov S.N., Lychenko N.M. Approximation of time series by a polymorphic wavelet network with feedbacks [Matematicheskie struktury i modelirovanie]. 2016.vol. 2.no.38. pp.16-26. (In Russian)

19. Verzunov S.N., Lychenko N.M. Multiwave polymorphic wavelet network for predicting geophysical time series [Problemy avtomatiki i upravlenija]. 2017.vol. 1.no.32. pp.78-87. (In Russian)

20. A. J. Zaslavski. Numerical Optimization with Computational Errors. Springer International Publishing. 2016. 304 p.

21. Hyndman, R.J. “Time Series Data Library”, http://data.is/TSDLdemo. Accessed on 14.12.2017.

22. Hipel K.W., McLeod A.I. Time Series Modelling of Water Resources and Environmental Systems. Elsevier, Amsterdam, 1994. 1012p.