Show TOC

Function documentationUnivariate Forecast Combined with MLR

 

For estimating the effects of price change events on the demand, a combined mode is available in forecasting. This means that the system executes univariate forecasting, and uses MLR forecasting to calculate the effect of MLR-independent variables (causal variables). The system cleans the effects of causal variables from the history before executing univariate forecasting, which allows for greater forecast accuracy.

Features

In the MLR for Univ. Fcst (MLR Profile for Univariate Forecasting) field in the univariate forecast profile, you can specify the MLR forecast profile that you want the system to use to calculate the effect of MLR-independent variables during univariate forecasting. The history key figure and its version of the MLR profile must match the history key figure in the univariate profile. It is not possible to use like profiles for forecasting with the new combined mode.

To use this function, you must activate the business function SCM-APO-FCS, Events and Outliers, Causals, ABC/XYZ Classification (SCM_APO_FORECASTING_1). For more information, see SCM-APO-FCS, Events and Outliers, Causals, ABC/XYZ Classification.

Activities

To use this function, on the SAP Easy Access screen, choose Start of the navigation path Advanced Planning and Optimization Next navigation step Demand Planning Next navigation step Environment Next navigation step Maintain Forecast Profiles End of the navigation path. On the Univariate Profile tab page, in the Combination with MLR screen area, in the MLR for Univ. Fcst field, enter the name of the MLR forecast profile that you want the system to use together with the univariate forecast profile.

If you have entered an MLR profile in the MLR for Univ. Fcst field, the system performs the following activities:

  1. It reads the demand history.

  2. It applies phase-in and phase-out profiles to the future and performs workday correction.

  3. It runs MLR forecasting, it calculates only coefficients and does not save results.

  4. It cleans the history from the effect of MLR-independent variables.

  5. It runs univariate forecasting on the basis of the cleaned history.

  6. It adds the effect of MLR-independent variables back to the history and forecast.

  7. It applies phase-in and phase-out profiles to the history, performs workday correction and outlier correction.

  8. It calculates the forecast error and saves the results.