Forecasting tutorial: forecasting based on actuals

In this part of the tutorial you create a forecast based on actuals.

Before you begin

This tutorial is meant to be followed in sequential order. Each part assumes that the previous part was completed.

Note: This part of the tutorial creates and saves forecast data. Unlike the preview you created in Forecasting tutorial: previewing results, when you click Forecast (as opposed to Preview), you modify the data on your Planning Analytics database.

Procedure

  1. Configure the same Exploration view of the Tutorial_<your initials> cube as in the previous section:
    • Year on the columns
    • Members on the rows
    • Version on the context area, with the Actual member displayed
  2. Select Member1, Member2 and Member3 on the row dimension.
    Member2 and Member3 selected in an exploration view
  3. Click the Forecast icon Forecast icon on the Exploration toolbar.
  4. Set the Forecast start period to 2014.
  5. Set the Forecast end period to 2019.
  6. Enable the Save statistical details as comments option.
  7. Click the Advanced tab on the Forecast panel.
  8. Scroll down to the section titled Where do you want to save the predicted values? and configure the section as shown. Remember, the Version dimension name and the Version hierarchy name are appended with your initials.
    Advanced forecast configuration
  9. Click Forecast.

    When the forecast is done, an informational message indicates that the AI driven forecast completed successfully.

    The data in the cube refreshes automatically when the forecast is complete, but the results are not immediately visible in the view. This is because the forecast is calculated using Actual data values, but the forecast values are saved to the Forecast member.

  10. Click the Version dimension tile on the context are of the view and then select the Forecast member.
    You can see that forecast values were written and saved to the cube for the selected row members and the years 2014 - 2019.
    View showing forecast values for selected members
    The Result summary box and the Details link on the Forecast panel display information about the predictive strength of each time series in the forecast. In this case, two of the time series have a High prediction strength, while the other has a Low prediction strength.
    Result summary box onthe Forecast panel

    Click Details to see more information about the prediction strength of each time series.

    Because you enabled the Save statistical details as comments when you configured the forecast, you can also view statistical details for the forecast directly in the Exploration view. The first cell in a forecast time series displays a blue triangle as an indicator that comments or statistical details are saved with the cell.

    Cell with a comment indicator
  11. Right-click the cell in the forecast with the comment indicator, then click Comments to see the statistical details for the forecast on the associated row member.
  12. Click Close to dismiss the comments.
  13. Click Save on the Exploration toolbar.
  14. Enter Forecast_tutorial_<your initials> as the view name.
  15. Click the Shared tab to save the Exploration view to the Planning Analytics database Shared folder. When a view with forecast data is saved to the Shared folder or as a public view on the Planning Analytics database, other analysts can explore the forecast you create.
  16. Click Save.