Time series decomposition analysis

Decompose a time series into trend, seasonal, and residual components.

Time series decomposition shows four decompositions for the measure in the selected time period:

• Observed: The measure values for the time series.

• Trend: The increasing or decreasing value as determined by the data model.

• Seasonality: The repeating short-term cycle in the series.

• Residual: Values not included in the trend. Residual shows where a model is missing more than it is finding.

You can optionally remove decompositions from the analysis.

Creating time series decomposition analysis

Do the following:

1. In Assets, click Analysis.

2. Drag and drop Time series decomposition into the sheet.

3. Click Add dimension and select a temporal dimension.

4. Click Add measure and a select a measure.

5. In the properties panel, under Analysis properties, optionally customize your analysis:

   • To change the number of period segments, after Period segments, adjust the slider.

   • To remove decompositions, after Decompositions, clear the decompositions you do not want in your analysis.

6. Optionally, in the properties panel, under Appearance, adjust the appearance of the analysis.