eigenCAR1: Function to compute eigenvalues and eigenvectors of CAR1...

In CARrampsOcl: Reparameterized and marginalized posterior sampling for conditional autoregressive models, OpenCL implementation

Description

Function to compute eigenvalues and eigenvectors of CAR1 structure matrix.

Usage

eigenCAR1( nr, nc )

Arguments

nr	integer; first dimension of structure matrix
nc	integer; second dimension of structure matrix

Value

values	eigenvalues
vectors	eigenvectors

Author(s)

Kate Cowles

References

Kunsch, H.R. (1994), "Robust priors for smoothing andimage restoriation," Annals of the Institute of Statistical Mathematics, 55, no. 1, 1-19.

Examples

## The function is currently defined as
function(nr,nc)
{

eigenrow <- eigenRW1(nr)
eigencol <- eigenRW1(nc)

list( values = kronecker( rep(1,nc), eigenrow$values) +
        kronecker(eigencol$values, rep(1,nr)) ,
      vectors = kronecker( eigencol$vectors, eigenrow$vectors) )

}

CARrampsOcl documentation built on May 2, 2019, 3:27 a.m.