8.2 Data description

We will use two simulated data sets from SMCRM package. This package consists of 7 data sets used in the book titled “Statistical Methods in Customer Relationship Management” by Kumar and Petersen. Specifically we will use customerRetentionTransactions and customerRetentionDemographics data sets. The first data consists of customer transactions information on 500 customers over 12 quarters. There are no missing values so every customer has 12 observations and accordingly there are 6,000 observations (500*12). The second data set contains demographic information on 500 customers. This information is time-invariant.

The data sets and their descriptions are as follows:

8.2.1 customerRetentionTransactions

Table 8.1: Variable Description for customerRetentionTransactions
Variable Description
customer customer number (from 1 to 500)
quarter quarter (from 1 to 12) where the transactions occurred
purchase 1 when the customer purchased in the given quarter and 0 if no purchase occurred in that quarter
order_quantity dollar value of the purchases in the given quarter
crossbuy number of different categories purchased in a given quarter
ret_expense dollars spent on marketing efforts to try and retain that customer in the given quarter
ret_expense_sq square of dollars spent on marketing efforts to try and retain that customer in the given quarter

8.2.2 customerRetentionDemographics

Table 8.2: Variable Description for customerRetentionDemographics
Variable Description
customer customer number (from 1 to 500)
gender 1 if the customer is male, 0 if the customer is female
married 1 if the customer is married, 0 if the customer is not married
income 1 if income <= $30,000;
2 if $30,000 < income <= $45,000;
3 if $45,000 < income <= $60,000;
4 if $60,000 < income <= $75,000;
5 if $75,000 < income <= $90,000;
6 if income > $90,000
first_purchase value of the first purchase made by the customer in quarter 1
loyalty 1 if the customer is a member of the loyalty program, 0 if not
sow share-of-wallet; the percentage of purchases the customer makes from the given firm given the total amount of purchases across all firms in that category
clv discounted value of all expected future profits, or customer lifetime value