scenarios: scenarios function (fully optimized with Rcpp) This function...

In APRScenario: Structural Scenario Analysis for Bayesian Structural Vector Autoregression Models

scenarios function (fully optimized with Rcpp) This function computes the mean and covariances to draw from the conditional forecast The actual draw is done in the simscen function

Description

scenarios function (fully optimized with Rcpp) This function computes the mean and covariances to draw from the conditional forecast The actual draw is done in the simscen function

Usage

scenarios(
  h = 3,
  path = NULL,
  obs = NULL,
  free_shocks = NULL,
  n_sample = NULL,
  data_ = NULL,
  g = NULL,
  Sigma_g = NULL,
  posterior = NULL,
  matrices = NULL
)

Arguments

h	forecast horizon
path	conditional path of observables
obs	position of observable(s)
free_shocks	position of non-driving shocks (NA if all driving)
n_sample	Number of draws to sample (<= n_draws)
data_	Optional matrix of data (default taken from matrices$Z). Note: the last observation in data_ is used as the starting point; it should not overlap with the scenario forecasting period.
g	Optional matrix of non-driving shocks
Sigma_g	Optional covariance matrix of non-driving shocks
posterior	Optional posterior object (default taken from calling environment)
matrices	Optional matrices object from gen_mats() (default taken from calling environment)

Value

list of mu_eps, Sigma_eps, mu_y, Sigma_y, big_b, big_M, draws_used

Examples

library(APRScenario)
data(NKdata)

# Minimal example with a toy specification
spec <- bsvarSIGNs::specify_bsvarSIGN$new(as.matrix(NKdata[,2:4]), p = 1)
posterior <- bsvars::estimate(spec, S = 10)  # Use small S for fast test
matrices<-gen_mats(posterior = posterior, specification = spec)
# and having posterior object
 scenario_result <- scenarios(h = 2, 
                              path = c(1.0, 1.1), 
                              obs = 1, 
                              free_shocks = NA, 
                              posterior = posterior, 
                              matrices = matrices)

APRScenario documentation built on Dec. 22, 2025, 1:06 a.m.