This dataset contains a recency-frequency matrix capturing the
discrete transaction behavior of 11,104 customers over 6 transaction
opportunities, summarized as a recency-frequency matrix and a vector
of annual transactions.
Usage
data(donationsSummary)
Arguments
Format
$rf.matrix
A matrix with 22 rows (for each possible
recency-frequency combination in 6 calibration period transaction
opportunities) and 4 columns: number of transactions during the calibration
period (“x”), recency in the calibration period
(“t.x”), number of transaction opportunities in the
calibration period (“n.cal”), and number of customers with
this recency-frequency combination in the calibration period
(“custs”).
$rf.matrix.holdout
A matrix with 15 rows (for each possible
recency-frequency combination in 5 holdout period transaction
opportunities) and 4 columns: number of transactions during the
holdout period (“x.star”), recency in the holdout period
(“t.x.star”), number of transaction opportunities in the
holdout period (“n.star”), and number of customers with the
recency-frequency combination in the holdout period
(“custs”).
$x.star
A vector with 22 elements, containing the number of
transactions made by each calibration period recency-frequency bin
in the holdout period. It is in the same order as $rf.matrix.
$annual.sales
A vector with 11 elements, containing the
number of transactions made by all customers in each time period
in both the calibration and holdout periods.
Data from “a major nonprofit organization located in the midwestern
United States that is funded in large part by donations from
individuals. In 1995 the organization “acquired” 11,104 first-time
supporters; in each of the following six years, these individuals
either did or did not support the organization.”
This dataset contains, for each possible in-sample recency/frequency combination
in the 1995 cohort, the number of customers and the number of
transactions they made during the validation period.
References
Fader, Peter S., Bruce G.S. Hardie, and Jen
Shang. “Customer-Base Analysis in a Discrete-Time
Noncontractual Setting.” Marketing Science 29(6),
pp. 1086-1108. 2010. INFORMS. http://www.brucehardie.com/papers/020/