Show TOC

 Forecast with Linear Regression Model

Purpose

With this process the system executes a forecast with the linear regression model. This is a forecast model that can be used for products with trend behavior. The model carries out a (simple) linear regression (least squares method).

Before calculating the forecast, the system carries out outlier correction . After calculating the forecast, the system calculates the standard deviation and the MAD and performs a trend limitation .

For further information about how the system chooses this model, see Automatic Model Selection .

Process

  1. The system takes the demand from the first x periods of the historical analysis period, and places a straight line through the historical values using the least squares method.

    You can specify x on the SAP Easy Access screen under Start of the navigation path Advanced Planning and Optimization Next navigation step Service Parts Planning Next navigation step Planning Next navigation step Forecast Next navigation step Forecast Profile End of the navigation path on the Model Parameter tab page in the Periods for Trend Line in LR Model parameter.

    You can enter the historical analysis period on the General tab page in the Historical Periods parameter.

  2. The system stretches this line over the next periods. You can define the number of periods on the General tab page in the Forecast Periods parameter.

  3. The system pushes the time window of the first x periods one period forward, and then carries out linear regression.

  4. The system also stretches this line over the next periods that you can define in the Forecast Periods parameter. You can see the result of this calculation on the SAP Easy Access screen under Start of the navigation path Advanced Planning and Optimization Next navigation step Service Parts Planning (SPP) Next navigation step Planning Next navigation step Forecasting Next navigation step Interactive Forecasting End of the navigation path in the Demand: Expost Forecast key figure.

  5. The system continues to push the time window forward by one period, calculate the linear regression, and continue the regression line until it arrives at the current period. At this point the ex-post forecast then goes over into the actual forecast.