a3-df.docx - Model Summary Specifications Growing Method...

This preview shows page 1 - 3 out of 9 pages.

Model Summary Specifications Growing Method CRT Dependent Variable v12 Independent Variables s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11 Validation Cross Validation Maximum Tree Depth 8 Minimum Cases in Parent Node 20 Minimum Cases in Child Node 10 Results Independent Variables Included s10, s7, s1, s9, s11, s6, s3, s8, s5, s4, s2 Number of Nodes 9 Number of Terminal Nodes 5 Depth 3 Risk Method Estimate Std. Error Resubstitution .050 .013 Cross-Validation .094 .017 Growing Method: CRT Dependent Variable: v12 Classification Observed Predicted 1 2 1 144 5 2 10 140 Overall Percentage 51.5% 48.5% Growing Method: CRT Dependent Variable: v12 1. Validation: Cross validation Maximum tree depth:8
Minimum cases for parent node: 20 Minimum cases for child node: 10 Growing method: CRT Impurity measure: Gini Index 2. The final tree has 9 nodes with 5 terminal nodes. 3. S10, S11 and S1 are the top three most important features with highest values to make decision in our data on splitting the tree. Independent Variable Importance Independent Variable Importance Normalized Importance s10 .331 100.0% s11 .224 67.7% s1 .185 55.9% s7 .145 43.9% s9 .132 40.0% s6 .103 31.0% s3 .087 26.3% s8 .046 14.0% s5 .045 13.7% s2 .038 11.4% s4 .034 10.3% 4. Increasing number of cases for parent to 30 and child to 15 complexity decreases.

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture