Exponential Smoothing

Large Scale Statistical Forecasting. Explanation of Exponential Smoothing.

Nobody can really look into the future. However modern statistical methods, econometric models and Business Intelligence software can indeed to some extent help companies forecast and estimate what is going to happen in the future.

The Exponential Smoothing model

The Exponential Smoothing (ESM) model uses a weighted average of past and current values, adjusting weight on current values to account for the effects of swings in the data, such as seasonality. Using an alpha term (between 0-1), you can adjust the sensitivity of the smoothing effects. ESM is often used on Large Scale Statistical Forecasting problems, because it is both robust and easy to apply.

ESM is a popular scheme to create a smoothed Time Series. Whereas in Single Moving Averages the past observations are weighted equally, Exponential Smoothing assigns exponentially decreasing weights if the observation gets older. In other words: recent observations are given more weight in forecasting than the older observations.

In the case of moving averages, the weights assigned to the observations are the same and are equal to 1/N. In Exponential Smoothing, however, there are one or more smoothing parameters which must be determined or estimated. These choices determine the weights assigned to the observations.