For questions 3 - 6

A large bank is interested in identifying the probability of fraudulent online purchases. The model below uses a variable, SimilarityScore, which considers similarity with past purchases, and an indicator variable for large transaction amounts. LargeTransaction (coded 1 if a transaction is greater than $1,000 and coded 0 if not) to explain the log-odds of fraudulent purchases (coded 1 if the transaction is fraudulent and coded 0 if not)

Ln(odds of purchase fraud) = 12 - 0.018*Similarity + 0.4*LargeTransaction

3.) If a purchase has a SimilarityScore of 600 and a transaction amount of $1,200, what are the estimated odds that the purchase is fraudulent?

Group of answer choices

1.17

2.20

2.01

1.49

4.95

4.) If a purchase has a SimilarityScore of 720 and a transaction amount of $500, what is the estimated probability that the purchase is fraudulent?

Group of answer choices

0.314

0.211

0.286

0.242

0.277

5.) Which of the following statements about the coefficient associated with LargeTransaction is true?

Group of answer choices

A purchase that is larger than $1,000 is 1.492 times less likely to be fraudulent than a purchase that is less than $1,000, holding SimilarityScore constant.

A purchase that is larger than $1,000 is twice as likely to be fraudulent than a purchase that is less than $1,000, holding SimilarityScore constant.

A purchase that is larger than $1,000 is 1.822 times more likely to be fraudulent than a purchase that is less than $1,000, holding SimilarityScore constant.

A purchase that is larger than $1,000 is 1.492 times more likely to be fraudulent than a purchase that is less than $1,000, holding SimilarityScore constant.

6.) HoldingLargeTransaction constant, how will the predicted odds of purchase fraud change for a purchase with a SimilarityScore of 875 compared to 610?

Odds are 90.37 percent less

Odds are 99.15 percent less

Odds are 98.26 percent less

Odds are 0.85 percent less

Predicted odds would increase