SimulateAMSS: Generate simulation objects under the AMSS framework.

Description Usage Arguments Details Value

Description

Produces an amss.sim object that contains the simulated data and can be used to derive ground truth about the scenario.

Usage

1
2
3
4
5
6
SimulateAMSS(time.n, nat.mig.module = DefaultNatMigModule,
  nat.mig.params = list(), media.names = character(),
  media.modules = rep(list(DefaultTraditionalMediaModule),
  length(media.names)), media.params = rep(list(list()), length(media.names)),
  sales.module = DefaultSalesModule, sales.params = list(), ping = max(10,
  floor(time.n/10)), names.agg.const = NULL, names.agg.sum = NULL)

Arguments

time.n

number of timepoints.

nat.mig.module

specifications of class seq.specs, to be used to create the non-actionable drivers module and then generate its variables

nat.mig.params

any parameter values to pass to nat.mig.module

media.names

character vector of unique names of all media modules. ex: c("tv", "search")

media.modules

list of functions that simulate the behavior of each marketing intervention.

media.params

list of parameter value lists for each media module.

sales.module

function that models the sales in the category.

sales.params

list of any parameter values to pass to the sales module

ping

the spacing between time points at which to print a message updating the user on simulation progress.

names.agg.const

character vector of names of variables to surface in the observed data, aggregated using the first entry, since they are constant over the hidden states. By default, if not specified, SurfaceData() will pick up variables with names containing "price" and/or "budget", and the "pop.total" variable.

names.agg.sum

character vector of names of variables to surface in the observed data, aggregated using the function sum(). By default, SurfaceData() will pick up variables with names matching "revenue", "profit", "sales", "volume", and/or "spend".

Details

Objects of class amss.sim contain the full output from running a simulation scenario. This output includes the observed data, the complete dataset generated during the simulation process, and the parameters passed to the simulation function in order to generate the simulated data. The observed data is meant to be used by modelers. The complete dataset can be useful for users who want a more complete understanding of the forces operating in a simulation scenario. The parameter list is essential for generating future datasets based on the same, or slightly modified, simulation settings in order to obtain ground truth about the simulation scenario.

Value

an object of class amss.sim, containing

data

the observed data.

data.full

the full dataset, as a list of data.tables. Each data.table contains the data at the end of a time interval, by by population segment (row) and variable (column).

params

the parameters used to generate the data.


google/amss documentation built on May 20, 2019, 5:05 p.m.