Using the new Predictive Forecast process allows you to use historical data already within the Prophix cube to forecast future results. Used to aid in planning, Predictive Forecast can be an excellent and faster alternative to the Infoflex process.
The Predictive Forecast process creates a forecast by taking the source leaf-level actuals of a specific time frame and then applies time-series-based algorithms. The data resulting from these algorithms then populates the selected target version and time period.
Predictive Forecast is best used for specific accounts/expenses that have historical seasonality in the data set. The more historical the data, the more reliable the forecast.
To get started on this, follow the instructions below:
- Open process manager, click on Insert and select the Predictive Forecast process
- In the General tab select the cube you wish to do the Predictive Forecast in
- In the Source/Target Members tab under Source data select the members for each dimension you wish to base the forecast on.
Under target members, select the version you want the forecasted data to be written to.
- In the Forecast Properties tab, you will be able to choose between the Best fit, ARIMA, or Holt-Winter’s algorithm. You could click on the question mark to the right of this option to see an explanation of these algorithms in the Prophix help site.
For the Forecast Start Date, select the time period the forecast data should start. This would usually be the first time period in your forecast.
For Forecast Duration, specify the number of periods for which the forecast should be generated for.