Forecast multiple scenarios from a fitted posterior — chlaa_forecast_scenarios_from

Selects posterior iterations once (shared across scenarios), simulates baseline and each scenario using the same draw indices and RNG seed streams, and returns both absolute forecasts and paired differences vs baseline.

Usage

chlaa_forecast_scenarios_from_fit(
  fit,
  pars = NULL,
  scenarios = NULL,
  baseline_name = "baseline",
  time = NULL,
  vars = c("inc_symptoms", "cum_symptoms", "cum_deaths"),
  include_cases = TRUE,
  obs_model = c("nbinom", "mean"),
  quantiles = c(0.025, 0.25, 0.5, 0.75, 0.975),
  n_draws = 100,
  burnin = 0.5,
  thin = 1,
  seed = 1,
  dt = 0.25,
  n_particles = 1,
  n_threads = 1,
  deterministic = FALSE,
  include_baseline_in_scenarios = TRUE
)

Arguments

fit: A `chlaa_fit` object.
pars: Baseline parameter list. If NULL uses `attr(fit, "start_pars")` else `chlaa_parameters()`.
scenarios: Scenarios to run (list of `chlaa_scenario`, named list of modify lists, or a grid data.frame).
baseline_name: Baseline scenario name (modify list may be empty).
time: Simulation times. If NULL uses `fit` data times.
vars: Model variables to summarise.
include_cases: Include predicted observed cases variable "cases".
obs_model: One of "nbinom" or "mean".
quantiles: Quantiles to compute.
n_draws: Number of posterior draws to use.
burnin: Burn-in proportion or integer.
thin: Thinning interval.
seed: Seed.
dt: Model time step.
n_particles: Particles per draw.
n_threads: Threads for dust2.
deterministic: Deterministic process model toggle (if supported).
include_baseline_in_scenarios: If TRUE, ensures baseline is included even if not passed.

Value

A tidy data.frame with columns: scenario, type, time, variable, mean, quantiles, n_samples.