Example: Two illustrative customers, with and without VIP services offered Purchases ($200) No Purchases ($0) Without VIP With VIP Without VIP With VIP Incremental VIP sales $ 100.00 $ 0.00 $ 0.00 $ 0.00 Profit on incremental sales $ 40.00 $ 0.00 $ 0.00 $ 0.00 VIP service charge $ 23.25 $ 0.00 $ 3.25 $ 0.00 Incremental net profit $ 16.75 $ 0.00 $ 3.25 $ 0.00 Use the Training dataset to complete the following tasks. Evaluate all questions on the 2000 holiday season sales (HOL_MVAL). 1. (20 points) If you offer VIP Service to all customers in the data, what would be your total cost? Total benefit? Are you better offering VIP Service to all customers or no customers? 2. (20 points) What is the Expected Value of Perfect Information? In other words, if you could make your decision about which customers to offer VIP Service based on already knowing their HOL_MVAL, how much incremental profit would VIP Service create? 3. (30 points) The purchases made during the previous holiday season (M_VAL99H) has the highest correlation with current year holiday purchases, so create a regression model with only this variable to predict HOL_MVAL. How would you use this model to decide who to offer VIP Service? If you used this model on the Training dataset, how much incremental profit could you generate? Hint: You will decide which customers to offer VIP to using your regression model estimate and then you will assess your performance on actual sales (HOL_MVAL) like questions #1 and #2. The difference is question 3 is imperfect information so your performance should be better than the no information case but will fall short of question 2, perfect information