Classification

Figure 1. Classification table
Classification table

The classification table shows the practical results of using the Naive Bayes model. For each case, the predicted response is Yes if that cases's model-predicted probability of Yes is greater than the model-predicted probability of No.

  • Cells on the diagonal are correct predictions.
  • Cells off the diagonal are incorrect predictions.

Of the cases in the training sample, 60 of the 124 people who previously defaulted are classified correctly. 352 of the 375 nondefaulters are classified correctly. Overall, 82.6% of the cases are classified correctly.

By comparison, 78.1% of the cases in the test sample were correctly classified. Note that the test sample classification results do not provide a true validation of the model, since the test data criterion is used to rank the subsets. Even so, this suggests that, overall, this model is correct about four out of five times.

Next