# Exponential and Holt-Winters Smoothing

**DOI:**https://doi.org/10.1007/978-3-642-04898-2_244

Exponential smoothing techniques are simple tools for smoothing and forecasting a time series (that is, a sequence of measurements of a variable observed at equidistant points in time). Smoothing a time series aims at eliminating the irrelevant noise and extracting the general path followed by the series. Forecasting means prediction of future values of the time series. Exponential smoothing techniques apply recursive computing schemes, which update the previous forecasts with each new, incoming observation. They can be applied online since they only use past observations which are already available at the corresponding point in time. Although exponential smoothing techniques are sometimes regarded as naive prediction methods, they are often used in practice because of their good performance, as illustrated e.g. in Chapter 4 of Makridakis et al. (1998). We speak of exponential smoothing techniques in plural because different variants exist which have been designed for different...

## References and Further Reading

- Bowerman B, O’Connell R, Koehler A (2005) Forecasting, time series, and regression, Thomson Books Cole, Belmont, CAGoogle Scholar
- Gardner ES (2008) Exponential smoothing: the state of the art. Int J Forecast 22:637–666Google Scholar
- Gelper S, Fried R, Croux C (2010) Robust forecasting with exponential and holt-winters smoothing. J Forecast 29(3):285–300zbMATHMathSciNetGoogle Scholar
- Hyndman RJ, Khandakar Y (2008) Automatic time series forecasting: the forecast package for R. J Stat Softw 27:1–22Google Scholar
- Hyndman RJ, Koehler AB, Ord JK, Snyder RD (2008) Forecasting with exponential smoothing: the state space approach. Springer-Verlag, BerlinGoogle Scholar
- Hyndman RJ, Koehler AB, Snyder RD, Grose S (2002) A state-space framework for automatic forecasting using exponential smoothing methods. Int J Forecast 18:439–454Google Scholar
- Makridakis S, Wheelwright S, Hyndman R (1998) Forecasting, methods and applications, 3rd edn. Wiley, Hoboken, NJGoogle Scholar
- Ord JK, Koehler AB, Snyder RD (1997) Estimation and prediction for a class of dynamic nonlinear statistical models. J Am Stat Assoc 92:1621–1629zbMATHGoogle Scholar