latent_to_niwishart: Transform latent variables to Normalised Inverse Wishart

In finnlindgren/multiprobit: Bayesian Multivariate Probit Regression

Transform latent variables to Normalised Inverse Wishart

Description

Transform latent iid N(0,1) to Normalised Inverse Wishart matrices

Transform a Normalised Inverse Wishart matrix to latent iid N(0,1) variables

Usage

latent_to_niwishart(x, V_chol, df, lower_chol = FALSE)

latent_from_niwishart(W_chol, V_chol, df, lower_chol = FALSE)

Arguments

x	A vector of latent variables identifying the Normalised Inverse Wishart matrix, length d(d-1)/2
V_chol	The Cholesky factor of the Inverse Wishart V parameter
df	The Wishart degrees of freedom
lower_chol	TRUE if lower triangular Cholesky factors are used (default = FALSE)
W_chol	The Cholesky factor of the Normalised Inverse Wishart matrix

Value

A list with components

W_chol	The Cholesky factor of the Normalised Inverse Wishart matrix
s	Vector of scaling values
B_chol	The inner Cholesky factor of the Barlett decomposition

When lower_chol is FALSE, ⁠W_chol = (B_chol \%*\% V_chol) \%*\% diag(s)⁠.

When lower_chol is TRUE, ⁠W_chol = s * (V_chol \%*\% B_chol)⁠.

A list with components

x	A vector of latent variables identifying the Normalised Inverse Wishart matrix, length d(d-1)/2
s	Vector of scaling values
B_chol	The inner Cholesky factor of the Barlett decomposition

Note

This is an internal non-exported function

This is an internal non-exported function

Author(s)

Finn Lindgren

Finn Lindgren

Examples

## Not run: 
if (interactive()) {
  # EXAMPLE1
}

## End(Not run)
## Not run: 
if (interactive()) {
  # EXAMPLE1
}

## End(Not run)

finnlindgren/multiprobit documentation built on June 14, 2025, 1:12 p.m.