Predictor Importance (linear models)

Figure 1. Predictor Importance view
Horizontal bar chart of predictor importance. Amount of coverage in thousands is ranked first.
  1. Click the Predictor Importance view thumbnail.

This view shows the predictors in the final model in rank order of importance. For linear models, the importance of a predictor is the residual sum of squares with the predictor removed from the model, normalized so that the importance values sum to 1.

Note that a predictor's rank order in importance is not necessarily the order in which it was added to the model; for example, the transformed Type of claim was added first to the model but is second in importance.

Next