You can generate forecast data by using the OLAP time-series forecasting engine. This type of statistical forecasting uses previous results to predict future performance.
To create a forecast:
In the data source, add the forecast time periods to the time dimension for the cube.
Create a new measure for the forecast data in the same cube as the historical data.
Create a new cube script or open an existing script.
Create a forecast step:
On the General tab of the cube script, click New Step, then select New Forecast Step. The New Forecast Step property pages are displayed.
Complete the General page. For the forecast method, select Automatic unless you have strong reasons to select a different method. Click Help for additional information about this page.
Keep the default values on the Advanced Settings page, unless you chose a custom forecasting method.
For a custom forecast, set the Forecast Approach and Data Filters parameters.
On the Status page, select the historical time periods and the forecast time periods. The forecast periods are the last ones in status, based on the number specified on the General page.
Save the forecast step, then save the cube script.
To execute the forecast step, right-click the cube script and choose Execute Cube Script.
View the forecast data in the measure viewer.
Copyright © 2003, 2007, Oracle. All rights reserved.