RDF: Method Seasonal Regression Does Not React to Sales Drop as Expected in Some Scenarios

(Doc ID 2420031.1)

Last updated on JULY 09, 2018

Applies to:

Oracle Retail Demand Forecasting - Version 15.0.1 and later
Oracle Retail Predictive Application Server - Version 15.0.1 and later
Information in this document applies to any platform.

Symptoms

The RDF method Seasonal Regression does not appear to be reacting to sales drop as expected.

The system chooses this method when the historical sales is less than 2 years past. If there have been significant shifts in the level of sales from one year to the next, the model learns that shift and appropriately weighs last year's data (keeping the same shape from last year, but adjusting its scale).

There is an observed case where the forecast is copied from last year to the current year without considering the "weights", even when having a clear tendency of low sales, comparing the last year period with the current year, where expectation is the forecast to have been lowered.

Changes

 

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms