• Login
  • Register
  • Search

Comparison of Forecasting Effect of SARIMA Model and Holt-Winters Smoothing Based on GDP

Chenghan Li


This study analyses ARIMA Model and Holt-Winters Smoothing. In this report, the GDP of Australia is taken as an example to illustrate the differences between these two models and to decide which model could show a better performance in term of forecasting. RMSE (Root Mean Square Error) is used as an indicator to compare the forecasting results of ARIMA Model and Holt-Winters Smoothing. After analysing, Holt-Winters Smoothing is found that could provide a more accurate result. This report provides people with basic ideas that how to use basic forecasting techniques to explore the future trends of some economic indicators.


SARIMA Model; Holt-Winters Smoothing; GDP

Full Text:



Landefeld, J. S., Seskin, E. P., & Fraumeni, B. M. Taking the pulse of the economy: Measuring GDP [J]. Journal of Economic Perspectives 2008; 22(2): 193-216.

Gelper, S., Fried, R., & Croux, C. Robust forecasting with exponential and Holt–Winters smoothing [J]. Journal of

forecasting 2010; 29(3): 285-300.

Holt, C. C. Forecasting seasonals and trends by exponentially weighted moving averages [J]. International journal of forecasting 2004; 20(1): 5-10.

Hyndman, R. J., & Athanasopoulos, G. Forecasting: principles and practice [J]. 2018.

Vijayakumar, N., & Vennila, S. A comparative analysis of forecasting reservoir infl ow using ARMA model and Holt winters exponential smoothening technique [J]. International Jour. of Innovation in Science and Mathematics 2016; 4(2): 85-90.

Chamlin, M. B. Crime and arrests: An autoregressive integrated moving average (ARIMA) approach [J]. Journal of Quantitative Criminology 1988; 4(3): 247-258.

DOI: http://dx.doi.org/10.18686/fm.v6i2.3303