bgnbd.ExpectedCumulativeTransactions: BG/NBD Expected Cumulative Transactions

In BTYD: Implementing BTYD Models with the Log Sum Exp Patch

Description

Calculates the expected cumulative total repeat transactions by all customers for the calibration and holdout periods.

Usage

bgnbd.ExpectedCumulativeTransactions(
  params,
  T.cal,
  T.tot,
  n.periods.final,
  hardie = TRUE
)

Arguments

params	BG/NBD parameters - a vector with r, alpha, a, and b, in that order. r and alpha are unobserved parameters for the NBD transaction process. a and b are unobserved parameters for the Beta geometric dropout process.
T.cal	a vector to represent customers' calibration period lengths (in other words, the "T.cal" column from a customer-by-sufficient-statistic matrix).
T.tot	end of holdout period. Must be a single value, not a vector.
n.periods.final	number of time periods in the calibration and holdout periods. See details.
hardie	if TRUE, use h2f1 instead of hypergeo.

Details

The function automatically divides the total period up into n.periods.final time intervals. n.periods.final does not have to be in the same unit of time as the T.cal data. For example: - if your T.cal data is in weeks, and you want cumulative transactions per week, n.periods.final would equal T.star. - if your T.cal data is in weeks, and you want cumulative transactions per day, n.periods.final would equal T.star * 7.

The holdout period should immediately follow the calibration period. This function assume that all customers' calibration periods end on the same date, rather than starting on the same date (thus customers' birth periods are determined using max(T.cal) - T.cal rather than assuming that it is 0).

Value

Vector of expected cumulative total repeat transactions by all customers.

See Also

bgnbd.Expectation

Examples

data(cdnowSummary)

cal.cbs <- cdnowSummary$cbs
# cal.cbs already has column names required by method

params <- c(0.243, 4.414, 0.793, 2.426)

# Returns a vector containing cumulative repeat transactions for 273 days.
# All parameters are in weeks; the calibration period lasted 39 weeks.
bgnbd.ExpectedCumulativeTransactions(params,
                                     T.cal = cal.cbs[,"T.cal"],
                                     T.tot = 39,
                                     n.periods.final = 273,
                                     hardie = TRUE)

BTYD documentation built on Nov. 18, 2021, 1:10 a.m.