Learn R Programming
lame (version 1.3.4)

sample_sigma_beta_cpp: Sample the AR(1) innovation sigma for each dynamic block

Description

Inverse-Gamma full conditional, one sigma per block.

Usage

sample_sigma_beta_cpp(
  beta_path,
  group_id,
  n_groups,
  Lambda_inv,
  rho_by_group,
  prior_shape,
  prior_scale
)

Value

Length n_groups vector of new sigma values.

Arguments

beta_path

(T x p_dyn) matrix.

group_id

Length p_dyn integer vector (1-based) of block IDs.

n_groups

Number of distinct block IDs.

Lambda_inv

(p_dyn x p_dyn) inverse of the (full) Lambda scale.

rho_by_group

Length n_groups vector of current rho values.

prior_shape

Length n_groups vector of IG shape parameters.

prior_scale

Length n_groups vector of IG scale parameters.