Calculates the log-likelihood of the (M)BG/CNBD-k model.
mbgcnbd.cbs.LL(params, cal.cbs)mbgcnbd.LL(params, x, t.x, T.cal, litt)
bgcnbd.cbs.LL(params, cal.cbs)
bgcnbd.LL(params, x, t.x, T.cal, litt)
A vector with model parameters k, r,
alpha, a and b, in that order.
Calibration period customer-by-sufficient-statistic (CBS)
data.frame. It must contain a row for each customer, and columns x
for frequency, t.x for recency , T.cal for the total time
observed, as well as the sum over logarithmic intertransaction times
litt. A correct format can be easily generated based on the complete
event log of a customer cohort with elog2cbs.
frequency, i.e. number of re-purchases
recency, i.e. time elapsed from first purchase to last purchase
total time of observation period
sum of logarithmic interpurchase times
For bgcnbd.cbs.LL, the total log-likelihood of the provided
data. For bgcnbd.LL, a vector of log-likelihoods as long as the
longest input vector (x, t.x, or T.cal).
Platzer Michael, and Thomas Reutterer (forthcoming)