Home / Statistical Tools / Analysis Tools / Time Series / Double Exponential Smoothing
Double Exponential Smoothing¶
Smooth and forecast a series that has a trend, using Holt's method.
Where to find it
QXL Stat Tools → Analysis Tools → Time Series → Double Exponential Smoothing
Worked example
The Double Exponential Smoothing walkthrough builds this analysis step by step from sample data you can copy straight into Excel, and states every number it produces.
All six Time Series walkthroughs are listed on the How-To page.
What it does¶
Double exponential smoothing tracks two components as it walks through the series: the current level and the current trend. Each new observation updates both, weighted by the two smoothing constants. Because it carries a trend component, it follows trending data better than a moving average and can project forecasts forward.
Options¶
The dialog is Time Series Smoothing, with a Data tab and an Options tab.
Data tab
- Excel / GroupBy: analyze flat Excel columns, or split by one or more GroupBy columns
- Data Columns: one or more numeric columns; each produces its own analysis
- Overlay Datasets on a Single Chart: combine the selected series into one chart
Options tab (Double Exp Smoothing Options)
- Level (Alpha): smoothing constant for the level. Default 0.2; allowed 0 to 1 (entries outside the range are clamped). Larger values react faster to changes but smooth less.
- Trend (Gamma): smoothing constant for the trend. Default 0.2; allowed 0 to 1.
- Number of Forecasts: how many periods to forecast past the data. Default 0; must be 0 or more.
Output¶
A worksheet named DblExp ... containing:
- A line chart with the Actual, Fitted, and (when forecasting) Forecast series over the observation Index
- A data table with Index, Actual, Smooth, Predicted, Error columns
- An Accuracy Measures table with MAPE, MAD, and MSD