It should be quite apparent that if you make adjustments to more than one level of a hierarchy the reconciliation becomes quite complex. Although several of the examples presented seemed involved, these examples had simple hierarchies consisting of only 3 levels and 4 end-items. Most users of Forecast Pro are working with thousands of items and much more complex hierarchies.
In developing the reconciliation routine for Forecast Pro the authors considered using several different algorithms. In many situations the choice of which algorithm to use would change the forecast values—sometimes significantly. None of the algorithms were “statistically superior” to the others—they were simply different.
Virtually all of the complexity (and differences between algorithms) disappears if you restrict yourself to making forecast adjustments to a single level of the hierarchy. Although we’ve stated this previously, it’s worth repeating:
If you restrict your overrides to a single level of the hierarchy, the reconciliation process is simple and intuitive. We strongly urge you to limit your overrides to a single level if possible.