bssm (version 0.1.8)

particle_smoother: Particle Smoothing

Description

Function particle_smoother performs filter-smoother or forward-backward smoother, using a either bootstrap filtering or psi-auxiliary filter with stratification resampling.

Usage

particle_smoother(object, nsim, ...)
# S3 method for gssm
particle_smoother(object, nsim,
  seed = sample(.Machine$integer.max, size = 1), ...)
# S3 method for ngssm
particle_smoother(object, nsim, filter_type = "bsf",
  seed = sample(.Machine$integer.max, size = 1), max_iter = 100,
  conv_tol = 1e-08, ...)
# S3 method for nlg_ssm
particle_smoother(object, nsim, filter_type = "psi",
  seed = sample(.Machine$integer.max, size = 1), max_iter = 100,
  conv_tol = 1e-08, iekf_iter = 0, ...)
# S3 method for sde_ssm
particle_smoother(object, nsim, L,
  seed = sample(.Machine$integer.max, size = 1), ...)

Arguments

object

Model.

nsim

Number of samples.

...

Ignored.

seed

Seed for RNG.

filter_type

Choice of particle filter algorithm. For Gaussian models, only option is "bsf" (bootstrap particle filter). In addition, for non-Gaussian or non-linear models, "psi" uses psi-particle filter, and for non-linear models options "ekf" (extended Kalman particle filter) is also available.

max_iter

Maximum number of iterations used in Gaussian approximation. Used psi-PF.

conv_tol

Tolerance parameter used in Gaussian approximation. Used psi-PF.

iekf_iter

If zero (default), first approximation for non-linear Gaussian models is obtained from extended Kalman filter. If iekf_iter > 0, iterated extended Kalman filter is used with iekf_iter iterations.

L

Integer defining the discretization level.