GWishart_BIPS_pairwise: BIPS algorithm for sampling from G-Wishart distribution

In NonDecompGraph: NonDecomposable Graph

Description

BIPS algorithm for sampling from G-Wishart distribution

Usage

GWishart_BIPS_pairwise(bG, DG, adj, C, burnin, nmc)

Arguments

bG	d.f.
DG	location
adj	adjacency matrix
C	initial precision matrix
burnin	number of MCMC burnins
nmc	number of saved samples

Details

BIPS algorithm for sampling from G-Wishart distribution with density: p(C) \propto |C|^{(bG-2)/2} exp(-1/2 tr(C DG)) where (1) bG : d.f. (2) DG: location (3) adj: adjacency matrix C: initial precision matrix;

Value

C	Samples from G-Wishart distribution. The result is 3D array with c(dim(C)[1],dim(C)[2], nmc ).
Sig	Inverse of C The result is 3D array with c(dim(C)[1],dim(C)[2], nmc ).

Author(s)

Hao Wang ; Sophia Zhengzi Li

References

Wang and Li (2011) "Efficient Gaussian Graphical Model Determination without Approximating Normalizing Constant of the G-Wishart distribution " http://www.stat.sc.edu/~wang345/RESEARCH/GWishart/GWishart.html

NonDecompGraph documentation built on May 2, 2019, 6:47 p.m.