Simulate Model Using Parameters with Maximum Log Posterior — simulate_with_max_posterior

Runs a model simulation using the parameter set corresponding to the highest posterior density (i.e., the maximum log posterior value) from the MCMC output. The function includes a prewarm period prior to the specified start_date to stabilize the system before collecting simulation results.

Usage

simulate_with_max_posterior_params(
  results,
  start_date,
  end_date,
  model,
  prewarm_years = 2,
  days_per_year = 360,
  mu_transform_A = NULL,
  mu_transform_C = NULL,
  covariate_matrix = NULL
)

Arguments

results: A list containing MCMC results. Must include log_posterior, param_inputs, and samples of estimated parameters.
start_date: Simulation start date (as Date or character string) for the output period.
end_date: Simulation end date (as Date or character string).
model: The model function to simulate. Should be compatible with data_sim().
prewarm_years: Number of years to simulate prior to start_date to allow system stabilization (default is 2).
days_per_year: Number of days in a simulation year (default is 360).
mu_transform_A: Optional transformation function for adult incidence (default is NULL).
mu_transform_C: Optional transformation function for child incidence (default is NULL).
covariate_matrix: Optional covariate matrix used in incidence transformation (default is NULL).

Value

A data frame containing model simulation results from start_date to end_date.

Examples

if (FALSE) { # \dontrun{
# Run simulation using max-posterior parameter set
sim_df <- simulate_with_max_posterior_params(
  results = results,
  start_date = "2021-01-01",
  end_date = "2021-12-31",
  model = data_sim
)
} # }