eigenRW1: Function to compute eigenvalues and eigenvectors of RW1...

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

Description

Function to compute eigenvalues and eigenvectors of RW1 structure matrix.

Usage

eigenRW1(n)

Arguments

n	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(n)
{
# compute eigenvals and eigenvects of RW1

w <-( pi/n) * ((n-1):0)
evals <-  2 * (1-cos(w) )

evects <- sin( outer( 1:n, w ) ) - sin( outer( (0:(n-1)), w))
evects[,n] <- rep(1,n)

evects <- apply( evects, 2, function(v) v / sqrt( sum(v^2) ) )

list( values = evals, vectors = evects)

}

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