Bookbinders Case 2

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 6

INTRODUCTION:

One of the principles of Database Marketing is you will lose money


sending a terrific piece of mail to a lousy list, but make money sending a
lousy piece of mail to a terrific list. With this case, you will learn to
evaluate DATABASE MARKETING CAMPAIGNS; will be exposed to customer
choice logit and regression models.
In this case, Bookbinders Book Club is considering two different methods
(i.e., customer choice logit model and linear regression model) to
understand the factors that influence customers order of the book titled
Art of History of Florence.
QUESTIONS:
Q1. Explain what does the statement in the Introduction (in bold and
italics) mean? Find one or two successful and unsuccessful company
examples of catalogue/mail marketing? Explain why some are successful
while others are not. Give one or two examples that are relevant to this
case.
Q2. Run the Customer Choice (Logit) Model. On the Diagnosis sheet, the
response row in Variable Averages table gives the average of each
variable for the whole sample (i.e., 1600 people). The Variable Averages
for Chosen Alternatives table gives the average of each variable for two
groups: those who choose to buy and not to buy the book. Explain
how the average values characterize each group and how group averages
are different from the whole sample.

Q3. Interpret the Confusion Matrix, for how many cases (i.e., respondents)
does the choice logit model predict their choice accurately? For each cell
in the matrix, give an exemplary respondent from the Estimation sheet.
Do you consider this as a successful estimation for the actual dataset?
Why?

As shown in above, the confusion matrix shows 80% accuracy. We find this by
adding the two bold numbers in each column (160+ 1120) and then divide by the
total number of respondents (1600).
Q4. Interpret the Coefficient Estimates (i.e., which variables affect
customer choice significantly?; Based on the significant variables, which of
them affect customer choice positively and negatively?) Please follow the
methodology we did in class and do not worry about the green and red
colors. In answering this question, do not just write down the list of
variables but try to explain these relationships in meaningful sentences as
if you are reporting it to your manager!

Q5. Run a linear regression model on the data. In the regression output,
find the Coefficients column and interpret the coefficients with their tstatistics (similar to the choice logit model output). Are the relationships
consistent between two models? Also, report the R-squared value in the
output and interpret the goodness-of-fit of the regression.

Shown above is the linear regression model. The r-squared value(highlighted) is .56
or 56% which does not show a good fit.
Q6. Based on the insights you gained, summarize the advantages and
limitations of the customer choice and regression models.

Figure 6:
4

You might also like