The Marketing Director of ACME Direct tested a new book title exactly one year ago and has decided, based on the results of the test, to promote this title to selected names from the database. Last month ACME Direct purchased, for the first time, new list enhancement data (age, income, marital status, home value, etc.) not previously on the customer database.
Using the saved sample from the original test promotion one year ago, the analyst is preparing to develop a regression model which will aid in predicting the type of customer most likely to order this particular book title. The Marketing Director has asked the analyst to append the new enhancement data to the sample in order to see if any of this “new enhancement data” will come into the regression equation.
Do you have any concerns regarding the marketing director’s request? If so, please clearly explain them. If not, please clearly explain why not.
Answer the following two questions related to regression modeling.
a) You are the senior product manager at AMCE Direct. Your analyst has built a regression model predicting a customer’s likelihood of ordering a product. You notice that the model contains two highly correlated variables which are causing multicollinearity problems. In questioning your analyst about this fact, he tells you the only way to rid the equation of this problem is to delete one of the two correlated variables. Do you agree with your analysts comments? Explain your answer fully.
b) What does it mean when a variable in your regression model has a high p-value associated with it? Fully explain your answ
Using the gains charts shown above the Senior Product Manager at ACME Direct will determine who to promote for his upcoming cookbook promotion. To ensure 5% profit-after-overhead for this campaign, the Senior Product Manager has determined he should not promote any group of customers with a response rate below 4.00%.
If the primary customer segment for the books product line (the universe the regression model was built on) represents 3,450,000 names:
a) How many names should the senior product manager promote? Please show how you calculated this.
b) What will be his expected number of orders in roll-out? Please show how you calculated this.
ACME Direct is planning to promote a new music package called “Rhythm and Blues” to its primary customer segment. Two logistic models were built predicting (1) the probability a customer will order the title, and (2) the probability a customer will not pay for the title if ordered. The package costs per thousand quantities for this promotion and the values associated with each customer action are shown below.
Marketing Assumptions for “Rhythm and Blues”:
Promotion cost per thousand (CPM) = $998
Value of a paid order = $25.79
Cost of an unpaid order = $7.45 (includes billing and lost product costs)
You score customer Smith on both logistic models. The values are shown below.
Customer Smith Logistic Regression Probability Scores:
Probability of ordering “Rhythm and Blues” .3548
Probability of not paying for “Rhythm and Blues” if ordered .0658
a) Customer smith has three courses of action he can take if promoted for this title. What are they?
1. ____________________________________________________________
2. ____________________________________________________________
3. ____________________________________________________________b) Determine customer Smith’s expected value to the corporation if promoted for this product using all information provided. Show all your work in determining this.
Sample Solution
