Comparing Bottom-up, Top-down, and Full Predictions

Full predictions, the default, predict all members on a form without regard to dimension hierarchies. With this method, Predictive Planning makes no assumption about the type of aggregation on the form.

Bottom-up predictions involve predicting members at the lowest levels of the dimension hierarchies and optionally rolling up the results to higher level summary members. This type of prediction requires that historical data be available for the lowest level members.

Top-down predictions involve predicting members at the summary levels of the dimension hierarchies and optionally spreading the results down to lower level members. This type of prediction is useful when historical data is not available for lower level members, or when top level predictions are being used to drive the results down to lower members.

Note:

Prediction results between full, bottom-up, and top-down methods should be close; however, predictions on lower level members are the most accurate because the individual trends and patterns of the data are preserved in the prediction process. If you are using top-down or full predictions and want to preserve the results at the summary-levels, ensure that the Planning business logic does not aggregate the results from lower level members.