Run Model Simulation with Pre-Warm Period

Extends time-varying covariate inputs and runs the model over a pre-warm period to allow system stabilization before the target simulation window.

Usage

run_simulation_with_prewarm(
  model,
  param_inputs,
  dates,
  prewarm_years = 2,
  days_per_year = 360
)

Arguments

model

A model object used for simulation (compatible with data_sim()).

param_inputs

A list of model parameters, including time-varying covariates.

dates

A vector of two dates: start and end of the target simulation window (as Date or string in "YYYY-MM-DD" format).

prewarm_years

Integer; number of years to pre-warm the model before the actual simulation window. Defaults to 2.

days_per_year

Integer; number of time steps per year (typically 360 or 365). Defaults to 360.

Value

A data frame of simulated results filtered to include only dates from the original start_date onward.

Details

Time-varying parameters are extended backward in time by repeating the first year's values for prewarm_years. The simulation is then run over this extended window, and the pre-warm results are discarded to avoid initialization artifacts.

Examples

if (FALSE) { # \dontrun{
sim_results <- run_simulation_with_prewarm(
  model = malaria_model,
  param_inputs = parameters,
  dates = c("2020-01-01", "2022-12-31"),
  prewarm_years = 2
)
} # }