Example of Single Exponential Smoothing

A toothpaste retailer collects sales data and the number of commercials that were broadcast during the previous 60 weeks. The retailer creates a time-series model for prediction of sales.

  1. Open the sample data, ToothpasteSales.MTW.
  2. Choose Stat > Time Series > Single Exp Smoothing.
  3. In Variable, enter Sales.
  4. Select Generate forecasts. In Number of forecasts, enter 6.
  5. Click OK.

Interpret the results

On the smoothing plot, the forecasts and line closely follow the data, especially at the end of the series. The retailer can be 95% confident that the sales will be between approximately 52 and 65 for the next 6 months.

Single Exponential Smoothing for Sales

Method Data Sales Length 101
Smoothing Constant α 0.951534
Accuracy Measures MAPE 7.9574 MAD 2.5674 MSD 11.3479
Forecasts Period Forecast Lower Upper 102 58.1889 51.8989 64.4790 103 58.1889 51.8989 64.4790 104 58.1889 51.8989 64.4790 105 58.1889 51.8989 64.4790 106 58.1889 51.8989 64.4790 107 58.1889 51.8989 64.4790

Single Exponential Smoothing Plot for Sales

By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy