Aggregate scenario outcomes across replicates — sce_aggregate

Function to aggregate outcomes from the replicate data of a scenario object. The use case of this function is to provide a compact representation of the model output, especially that of stochastic models.

Usage

sce_aggregate_outcomes(
  x,
  grouping_variables,
  measure_variables,
  summary_functions = c("mean", "sd")
)

Arguments

x: A scenario object with data prepared.
grouping_variables: The variables that should be used to group the outcomes of interest. Examples include demographic and susceptibility groups.
measure_variables: The outcomes of interest which are summarised over scenario replicates, and by all variables in the grouping_variables. Examples include p_infected from the output of finalsize::final_size().
summary_functions: The summary function names to apply to the measure variables, passed as strings, i.e., "mean" rather than simply mean.

Value

A data.table with the outcomes of interest (measure variables) summarised using the summary functions, by each grouping variable.

Examples

# create a scenario
scenario_pandemic_flu <- scenario(
  model_function = "finalsize::final_size",
  parameters = make_parameters_finalsize_UK(),
  replicates = 3 # note extra replicates
)

# run scenario
scenario_pandemic_flu <- run_scenario(scenario_pandemic_flu)

# peek at outcome to see column names
sce_peek_outcomes(scenario_pandemic_flu)
#>       demo_grp       susc_grp susceptibility     p_infected      replicate 
#>    "character"    "character"      "numeric"      "numeric"      "integer" 

# aggregate outcome by demographic group
sce_aggregate_outcomes(
  x = scenario_pandemic_flu,
  grouping_variables = c("demo_grp"),
  measure_variables = c("p_infected"),
  summary_functions = c("mean", "min", "max")
)
#> Key: <demo_grp>
#>    demo_grp p_infected_mean p_infected_min p_infected_max
#>      <char>           <num>          <num>          <num>
#> 1:      40+       0.4588871      0.4588871      0.4588871
#> 2:   [0,20)       0.6544866      0.6544866      0.6544866
#> 3:  [20,40)       0.5750030      0.5750030      0.5750030