Predictive Modeling with SAS Enterprise Miner: Practical Solutions for Business Applications, Third Edition by Sarma Kattamuri S

Author:Sarma, Kattamuri S. [Неизв.]
Language: eng
Format: epub
Publisher: SAS Institute

You can see a plot of the actual and predicted means for the training and validation data sets in the Results window of the Decision Tree Node .

4.5.1 Summary of the Regression Tree Model to Predict Risk

Although the regression tree shown in Displays 4.28A-4.28C is developed from simulated data, it provides a realistic illustration of how this method can be used to identify the profiles of high-risk groups.

We used the regression tree model developed in this section to group the records (in this case, the customers) in the data set into seven disjoint groups—one for each of the seven leaf nodes or terminal nodes on the tree—as shown in Display 4.30. The tree model also reports the mean of the target variable for each group, calculated from the Training data set. In this example, these means are used as predictions of the target variable loss frequency for each of the ten groups. Since the mean of the target variable for each group can be used for prediction, I refer to it as Predicted Loss Frequency in Display 4.30. The predicted loss frequency of a group reflects the level of risk associated with the group. The groups are shown in decreasing order of their risk levels. Display 4.30 also shows the profiles of the different risk groups, that is, their characteristics with regard to previous violations, credit score, and age.

Display 4.30

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.