mbgcnbd.ConditionalExpectedTransactions: (M)BG/CNBD-k Conditional Expected Transactions

In BTYDplus: Probabilistic Models for Assessing and Predicting your Customer Base

Description

Uses (M)BG/CNBD-k model parameters and a customer's past transaction behavior to return the number of transactions they are expected to make in a given time period.

Usage

mbgcnbd.ConditionalExpectedTransactions(params, T.star, x, t.x, T.cal)

bgcnbd.ConditionalExpectedTransactions(params, T.star, x, t.x, T.cal)

Arguments

params	A vector with model parameters k, r, alpha, a and b, in that order.
T.star	Length of time for which we are calculating the expected number of transactions.
x	Number of repeat transactions in the calibration period T.cal, or a vector of calibration period frequencies.
t.x	Recency, i.e. length between first and last transaction during calibration period.
T.cal	Length of calibration period, or a vector of calibration period lengths.

Value

Number of transactions a customer is expected to make in a time period of length t, conditional on their past behavior. If any of the input parameters has a length greater than 1, this will be a vector of expected number of transactions.

References

(M)BG/CNBD-k: Reutterer, T., Platzer, M., & Schroeder, N. (2020). Leveraging purchase regularity for predicting customer behavior the easy way. International Journal of Research in Marketing. doi: 10.1016/j.ijresmar.2020.09.002

Examples

## Not run: 
data("groceryElog")
cbs <- elog2cbs(groceryElog)
params <- mbgcnbd.EstimateParameters(cbs, k = 2)
# estimate transactions for next 12 weeks
xstar.est <- mbgcnbd.ConditionalExpectedTransactions(params,
  T.star = 12, cbs$x, cbs$t.x, cbs$T.cal)
head(xstar.est) # expected number of transactions for first 6 customers
sum(xstar.est) # expected total number of transactions during holdout

## End(Not run)

BTYDplus documentation built on Jan. 21, 2021, 5:10 p.m.