Measures of FitDemand Planning offers you a number of ways to measure how well an MLR (multiple linear regression) model fits the available data:
You view these measures in the data view
Multiple Linear Regression
of the demand planning desktop.
If you run the forecast with mass processing, use alerts to see when the upper or lower limit of the measure of fit has been exceeded. You can view alerts on the demand planning desktop or in the Alert Monitor.
You define the upper and lower limits of the measures of fit in a diagnostic group. You create a diagnostic group in the MLR profile. You can change these limits in the data view
Multiple Linear Regression
of the demand planning desktop.
While setting up your MLR model, you should remove independent variables from the model, if, based on the chosen measures of fit, they do not make a significant contribution to explaining the dependent variable. See also R square and Adjusted R square .