2
$\begingroup$

Decision Tree

I have found Misclassification rates for all the leaf nodes.

  1. samples = 3635 + 1101 = 4736, class = Cash, misclassification rate = 1101 / 4736 = 0.232.

  2. samples = 47436 + 44556 = 91992, class = Cash, misclassification rate = 44556 / 91992 = 0.484.

  3. samples = 7072 + 15252 = 22324, class = Credit Card, misclassification rate = 7072 / 22324 = 0.317.

  4. samples = 1294 + 1456 = 2750, class = Credit Card, misclassification rate = 1294 / 2750 = 0.470.

  5. samples = 7238 + 22295 = 29533, class = Credit Card, misclassification rate = 7238 / 29533 = 0.245.

I'm finding it difficult to find AUC value from here. Please help me out with this. I will be grateful.

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

A ROC curve is created by doing the following.

  1. Make predictions over a range of values (such as the probability of membership in a category).

  2. Apply a threshold to bin those predictions into two categories.

  3. Calculate the sensitivity and specificity when that threshold is applied.

  4. Vary the threshold all over to get many different values of sensitivity and specificity.

Your predictions seem to be the categories rather than over a range of values. Consequently, there is not much that you can do with a ROC curve. (You can do a little bit, but not much.) However, your software package likely has a function or method that allows you to access the predictions made over a range of values. Then you would be able to calculate a nice ROC curve.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.