Optimizing Prices This exercise is based on "Fjord Motors" case. Use the dataset "HW3 FjordMotor_Data.xis" to analyze and answer the questions below. Mike Selvage and Dave Tobin are consultants working for a data analytics startup based in Austin TX. They are presently advising Fjord Motor's B2B sales team in revamping their bidding strategy for fleet sales auctions for their popular model Coronet Elizabeth. In the past quarter, Fjord Motors has experimented with about 4,000 bid prices (using managerial intuition and heuristics) and Messrs. Selvage and Tobin have been given access to this data that has the bid price, auction outcome and the order quantity. The MSRP for Coronet Elizabeth during the quarter was $25,000 and the total cost per unit can be assumed to be $15,000. Part 1A. Using the attached datasheet, they fit a two-parameter logit model that best estimates the probability of wining each bid as a function of discount from the list price, assuming a single per unit price will be offered for each bid. The model they fit is: P(P) = exp (a+bp) 1+exp (a+bp) Where "a" and "b" are the parameters to be estimated and "p" is the probability of winning the bid. The price p is expressed as a fraction of the MSRP of the $25,000; that is, if Fjord bid $20,000 per unit, p would be equal to 0.8. (1) What are the values of "a" and "b" (obtained by running a logit model)? (2) What is the optimal price Fjord, assuming it is going to offer a single price for each bid? (3) What is the total expected contribution, had Fjord used the price in (2) to bid in all 4,000 auctions? (4) How does the answer in (3) compare with the actual total contribution received by Fjord by participating in the 4,000 auctions