A large number of insurance records are to be examined to develop a model for predicting fraudulent claims. Of the claims in the historical database, 1% were judged to be fraudulent (class 1). A sample database is taken to develop a model, and oversampling is used to provide a balanced sample in light of the very low response rate. When applied to this sample database (total number of records, N = 800), the model ends up correctly classifying 310 frauds, and 270 non-frauds. It misses 90 frauds, and classified 130 records incorrectly as frauds when they were not.

a) If the positive sample number is fixed (400), what is the total number of true negative records that should be in the original non-oversampled database?

the sample ratio is 1:99 (fraudulent vs. non-fraudulent, positive vs.negative)

b) If the positive sample number is fixed (400), what is the total number of records that should be in the original non-oversampled database?

c) If the positive sample number is fixed (400), what is the adjusted misclassification rate (error rate) that should be in the original non-oversampled database?

d) If the positive sample number is fixed (400), what is the total number of negative records that should be in the original non-oversampled database?

e) If the positive sample number is fixed (400), what is the adjusted positive response rate that should be in the original non-oversampled database? (Positive responses by the model means the number of records are classified as positive (here fraudulent) by the model.) The sample ratio is 1:99 (fraudulent vs. non-fraudulent, positive vs. negative)

f) If the positive sample number is fixed (400), what is the total number of false postive records that should be in the original non-oversampled database?

the sample ratio is 1:99 (fraudulent vs. non-fraudulent, positive vs. negative)