rGibbsHurdle: Sample from a multivariate hurdle model

In amcdavid/HurdleNormal: 'Estimation and sampling for multi-dimensional Hurdle models on a Normal density with applications to single-cell co-expression'

Sample from a multivariate hurdle model

Description

Sample from a multivariate hurdle model

Usage

rGibbsHurdle(G, H, K, Nt, burnin = 500, thin = 0.1, tol = 5e-04, Nkeep = 500)

Arguments

G	symmetric discrete interaction matrix
H	unstructured location matrix
K	Symmetric positive definite conditional precision matrix
Nt	Number of unthinned samples, including burnin
burnin	how many samples to discard from burn in
thin	how many samples to thin
tol	Numeric tolerance for zero
Nkeep	(optional) number of samples, post-burnin and thinning

Value

matrix of (Nt-burnin)*thin samples

Examples

G = matrix(c(-15, 1, 0,
1, -15, 1.5,
0, 1.5, -15), nrow=3)
H = diag(5, nrow=3)
K = diag(1, nrow=3)
y = rGibbsHurdle(G, H, K, 2000, thin = .2, burnin=1000)

amcdavid/HurdleNormal documentation built on May 14, 2022, 11:12 p.m.