A Study on Internet Traffic Forecasting by Combined Forecasts

결합예측 방법을 이용한 인터넷 트래픽 수요 예측 연구

  • Received : 2015.11.02
  • Accepted : 2015.11.23
  • Published : 2015.12.31


Increased data volume in the ICT area has increased the importance of forecasting accuracy for internet traffic. Forecasting results may have paper plans for traffic management and control. In this paper, we propose combined forecasts based on several time series models such as Seasonal ARIMA and Taylor's adjusted Holt-Winters and Fractional ARIMA(FARIMA). In combined forecasting methods, we use simple-combined method, MSE based method (Armstrong, 2001), Ordinary Least Squares (OLS) method and Equality Restricted Least Squares (ERLS) method. The results show that the Seasonal ARIMA model outperforms in 3 hours ahead forecasts and that combined forecasts outperform in longer periods.


Fractional Seasonal ARIMA;adjusted Holt-Winters;internet traffic;combined forecasting


  1. Aksu, C. and Bunter, S. (1992). An empirical analysis of the accuracy of SA, OLS, ERLS and NRLS combination forecasts, International Journal of Forecasting, 8, 27-43.
  2. Armstrong, J. S. (2001). Combining forecasting, International Series in Operations Research and Manage-mert Science, 417-440.
  3. Au, S. T., Ma, G. Q. and Yeung, S. N. (2011). Automatic forecasting of double seasonal time series with applications on mobility network traffic prediction, In 2011 Joint Statistical Meetings, July.
  4. Box G. E. P. and Jenkins, G. (1994). Time Series Analysis: Forecasting and Control, Prentice Hall.
  5. Engle, R. F. (1982). Autoregressive conditional heteroskedasticity with estimates of the variance of U. K. Inflation, Econometrica, 50, 987-1008.
  6. Ha, M. and Kim, S. (2008). Internet traffic forecasting using power transformation heteroscedastic time series models, The Korea Journal of Applied Statistics, 21, 1037-1044.
  7. Jiang, J. and Papavassiliou, S. (2006). Enhancing network traffic prediction and anomaly detection via statistical network traffic separation and combination strategies, Computer Communications, 29, 1627-1638.
  8. Kim, S. (2007). Time series models for performance evaluation of network traffic forecasting, The Korea Journal of Applied Statistics, 20, 219-227.
  9. Kim, S. (2011). Forecasting internet traffic by using seasonal GARCH models, Journal of Communications and Network, 13, 621-624.
  10. Liu, J., Shu, Y., Zhang, L. and Xue, F. (1999). Traffic modeling based on FARIMA models, In IEEE Canadian Conference on Electrical and Computer Engineering, 621-624.
  11. Shu, Y. T., Yu, M. F., Liu, J. K. and Yang, O. W. W. (2003). Wireless traffic modeling and prediction using seasonal ARIMA models, Communications, 2003. ICC'03. IEEE International Conference on. IEEE, 3, 1675-1679.
  12. Sohn, H. and Kim, S. (2013). A study on the short term internet traffic forecasting models on long-memory and heteroscedasticity, The Korea Journal of Applied Statistics, 26, 1053-1061.
  13. Taylor, J. W. (2003). Short-term electricity demand forecasting using double seasonal exponential smoothing, Journal of the Operational Research Society, 54, 799-805.
  14. Tikunov, D. and Nishimura, T. (2007). Traffic prediction for mobile network using Holt-Winter's exponential smoothing, Telecommunications and Computer Networks, 2007. SoftCOM 2007. 15th International Conference, 1-5.
  15. Yu, Y. H., Wang, J., Song, M. N. and Song, J. D. (2010). Network traffic prediction and result analysis based on seasonal ARIMA and correlation coefficient, 2010 International Conference on Intelligent System Design and Engineering Application,
  16. Zhou, B., He, D., Sun, Z. and Ng, W. H. (2005). Network traffic modeling and prediction with ARIMA/GAR CH, In Proceeding of HET-NETs Conference, 1-10.


Supported by : 한국연구재단