mcmc.PlotTrackingCum: Tracking Cumulative Transactions Plot for Pareto/GGG,...

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

Tracking Cumulative Transactions Plot for Pareto/GGG, Pareto/NBD (HB) and Pareto/NBD (Abe)

Description

Plots the actual and expected cumulative total repeat transactions by all customers for the calibration and holdout periods, and returns this comparison in a matrix.

Usage

mcmc.PlotTrackingCum(
  draws,
  T.cal,
  T.tot,
  actual.cu.tracking.data,
  xlab = "Week",
  ylab = "Cumulative Transactions",
  xticklab = NULL,
  title = "Tracking Cumulative Transactions",
  ymax = NULL,
  sample_size = 10000,
  covariates = NULL,
  legend = c("Actual", "Model")
)

Arguments

draws	MCMC draws as returned by *.mcmc.DrawParameters
T.cal	A vector to represent customers' calibration period lengths (in other words, the T.cal column from a customer-by-sufficient-statistic matrix). Considering rounding in order to speed up calculations.
T.tot	End of holdout period. Must be a single value, not a vector.
actual.cu.tracking.data	A vector containing the cumulative number of repeat transactions made by customers for each period in the total time period (both calibration and holdout periods).
xlab	Descriptive label for the x axis.
ylab	Descriptive label for the y axis.
xticklab	A vector containing a label for each tick mark on the x axis.
title	Title placed on the top-center of the plot.
ymax	Upper boundary for y axis.
sample_size	Sample size for estimating the probability distribution. See mcmc.ExpectedCumulativeTransactions.
covariates	(optional) Matrix of covariates, for Pareto/NBD (Abe) model, passed to abe.GenerateData for simulating data.
legend	plot legend, defaults to 'Actual' and 'Model'.

Details

The expected transactions need to be sampled. Due to this sampling, the return result varies from one call to another. Larger values of sample_size will generate more stable results.

Value

Matrix containing actual and expected cumulative repeat transactions.

See Also

mcmc.PlotTrackingInc mcmc.ExpectedCumulativeTransactions elog2cum

Examples

## Not run: 
data("groceryElog")
cbs <- elog2cbs(groceryElog, T.cal = "2006-12-31")
cum <- elog2cum(groceryElog)
param.draws <- pnbd.mcmc.DrawParameters(cbs)
mat <- mcmc.PlotTrackingCum(param.draws,
  T.cal = cbs$T.cal,
  T.tot = max(cbs$T.cal + cbs$T.star),
  actual.cu.tracking.data = cum)

## End(Not run)

mplatzer/BTYDplus documentation built on April 9, 2024, 3:11 a.m.