A national bank has developed a predictive model for identifying customers who are more likely to accept a credit card offer. If a customer is predicted to accept the credit card offer, he or she is classified into Class 1; otherwise, he or she is classified into Class 0. The model is used to make predictions on a sample of 100 customers in the validation data set. The accompanying data file includes the customer sample, their actual class membership, and predicted Class 1 probability.

Click here for the Excel Data File

a-1. Specify the predicted class membership for the validation data set using the cutoff value of 0.25. Produce a confusion matrix.

a-2. Specify the predicted class membership for the validation data set using the cutoff value of 0.50. Produce a confusion matrix.

a-3. Specify the predicted class membership for the validation data set using the cutoff value of 0.75. Produce a confusion matrix.

b-1. Compute the misclassification rate, accuracy rate, sensitivity, precision, and specificity of the classification model for the cutoff value of 0.25.

Note: Round your final answers to 2 decimal places.

b-2. Compute the misclassification rate, accuracy rate, sensitivity, precision, and specificity of the classification model for the cutoff value of 0.50.

Note: Round your final answers to 2 decimal places.

b-3. Compute the misclassification rate, accuracy rate, sensitivity, precision, and specificity of the classification model for the cutoff value of 0.75.

Note: Round your final answers to 2 decimal places.

c-1. Create a cumulative lift chart for the classification model. At 60 cases, what is the cumulative response using the sorted predicted values?

c-2. Create a decile-wise lift chart for the classification model. What is the lift value of the first decile?

d. What is the lift that the classification model provides if 20% of the observations are selected by the model compared to randomly selecting 20% of the observations?

Note: Round your final answer to 2 decimal places.

e. What is the lift that the classification model provides if 50% of the observations are selected by the model compared to randomly selecting 50% of the observations?

Note: Round your final answer to 2 decimal places.