bssm (version 1.1.7-1)

importance_sample: Importance Sampling from non-Gaussian State Space Model

Description

Returns nsim samples from the approximating Gaussian model with corresponding (scaled) importance weights. Probably mostly useful for comparing KFAS and bssm packages.

Usage

importance_sample(model, nsim, use_antithetic, max_iter, conv_tol, seed, ...)
# S3 method for nongaussian
importance_sample(
  model,
  nsim,
  use_antithetic = TRUE,
  max_iter = 100,
  conv_tol = 1e-08,
  seed = sample(.Machine$integer.max, size = 1),
  ...
)

Arguments

model

Model of class bsm_ng, ar1_ng svm, ssm_ung, or ssm_mng.

nsim

Number of samples (positive integer).

use_antithetic

Logical. If TRUE (default), use antithetic variable for location in simulation smoothing. Ignored for ssm_mng models.

max_iter

Maximum number of iterations as a positive integer. Default is 100 (although typically only few iterations are needed).

conv_tol

Positive tolerance parameter. Default is 1e-8. Approximation is claimed to be converged when the mean squared difference of the modes of is less than conv_tol.

seed

Seed for the random number generator (positive integer).

...

Ignored.

Examples

# NOT RUN {
data("sexratio", package = "KFAS")
model <- bsm_ng(sexratio[, "Male"], sd_level = 0.001, 
  u = sexratio[, "Total"],
  distribution = "binomial")

imp <- importance_sample(model, nsim = 1000)

est <- matrix(NA, 3, nrow(sexratio))
for(i in 1:ncol(est)) {
  est[, i] <- Hmisc::wtd.quantile(exp(imp$alpha[i, 1, ]), imp$weights, 
    prob = c(0.05,0.5,0.95), normwt=TRUE)
}

ts.plot(t(est),lty = c(2,1,2))

# }