Ex Post Forecast is generated by the system during the statistical forecast calculation and only appears in historical periods. It shows you how the system is interpreting the demand history as it processes it. In this sense, it shows you how the system is "learning" the history. All forecast error metrics are calculated by comparing the ex post fcst to the demand history, so lower forecast error numbers mean the forecast profile assigned to the selection fits the history better.
In our implementation, we have included a key figure in the planning area for the ex post forecast so that the planners could see the comparison of ex post to history in the dataview graph. This lets them see at a glance how well the model traces the history.
You may also want to include a key figure for Corrected History in the planning area and in a dataview. The system actually copies history to 9AVCORHIST at runtime to use as the basis of the statistical forecast, and the results of background processes like Lifecycle Planning (Like, phase-in and phase-out profiles) and promotion planning show up here -- this makes it easier to trace what the system is doing and how it arrived at the statistical forecast.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.