BTYDplus (version 0.7.2)

mcmc.pmf: Probability Mass Function for Pareto/GGG, Pareto/NBD (HB) and Pareto/NBD (Abe)

Description

Return the probability distribution of purchase frequencies for a random customer in a given time period, i.e. \(P(X(t)=x)\). This is estimated by generating sample_size number of random customers that follow the provided parameter draws. Due to this sampling, the return result varies from one call to another.

Usage

mcmc.pmf(draws, t, x, sample_size = 10000)

Arguments

draws

MCMC draws as returned by pnbd.mcmc.DrawParameters, pggg.mcmc.DrawParameters or abe.mcmc.DrawParameters

t

Length of time for which we are calculating the expected number of transactions. May also be a vector.

x

Number of transactions for which probability is calculated. May also be a vector.

sample_size

Sample size for estimating the probability distribution.

Value

\(P(X(t)=x)\). If either t or x is a vector, then the output will be a vector as well. If both are vectors, the output will be a matrix.

Details

This is estimated by generating sample_size number of random customers that follow the provided parameter draws. Due to this sampling, the return result varies from one call to another.

See Also

bgcnbd.pmf pnbd.pmf

Examples

# NOT RUN {
cbs <- cdnow.sample()$cbs # load CDNow summary data
param.draws <- pnbd.mcmc.DrawParameters(cbs, 
  mcmc = 200, burnin = 100, thin = 20, chains = 1) # short MCMC runs for demo purposes
mcmc.pmf(param.draws, t = 52, x = 0:6)
mcmc.pmf(param.draws, t = c(26, 52), x = 0:6)
# }