1. Create a classification tree model for predicting whether the employees will be promoted into a management role after 10 years with the company. a. Select the best-pruned tree for scoring and display the full-grown, best- pruned, and minimum error trees. b. How many leaf nodes, are in the best-pruned tree and minimum error tree? c. What are the predictor variable and split value for the root node of the best-pruned tree? 2. Describe the rules produced by the best-pruned tree. 3. Give and explain the following of the best-pruned tree on the test data: a. accuracy rate b. sensitivity c. specificity d. precision

4. Display the a. cumulative lift chart b. the decile-wise lift chart c. and the ROC curve 5. Does the classification model outperform the baseline model? a. Why or why not. b. What is the area under the ROC curve (or AUC value of the model)? 6. Which is the most important predictor variable? 7. Score the new cases in the HR_Score worksheet using the best-pruned tree. a. What is the probability of the first new employee being promoted within 10 years according to your model? b. How many new employees in the data set will likely be promoted within 10 years based on a cutoff probability value of 0.5? The PowerPoint presentation shall contain (at a minimum) the following slides from the Analytic Solver Excel data output and graphs. 1. Classification tree model. 2. Rules derived from best-pruned tree. 3. Test data o accuracy rate o sensitivity o specificity o precision 4. Display o the cumulative lift chart o the decile-wise lift chart o and the ROC curve 5. Portion of Record table