Ex-Post Forecast 

Definition

A forecast that is run in past periods for which actual demand history is also available. The system calculates the forecast accuracy measurements by comparing the differences between the actual values and the ex-post values. For more information, see Forecast Accuracy Measurements.

Use

If more historical values are available than the system needs to initialize the model, an ex-post forecast is carried out automatically as follows:

●      The historical values are divided into two groups: the first group with the older values is used for initialization; the more recent values in the second group are used to carry out an ex-post forecast (see the figure below).

●      The basic value, the trend value, the seasonal index, and the mean absolute deviation (MAD) are modified in every ex-post period. These values are used to calculate the forecast results in the future.

●      The system calculates the error total in the ex-post horizon. The error total is the sum of the differences between the actual (historical) values and the planned (in this instance, the ex-post forecast) values.

●      The ex-post forecast results are stored in the key figure you predefined for this purpose in Customizing.

 

Other points to consider in connection with the ex-post forecast are:

●      It is not possible to influence the length of the initialization and ex-post horizons.

●      An ex-post forecast is carried out for all forecast strategies other than 13, 14, 60, 70 (the manual correction phase), 80 and 94.

●      For more information on the role of the ex-post forecast in forecast strategies 12 and 23, see Automatic Adaptation of the Alpha Factor.