sommer (version 1.9)

S.mat: Spatial relationship matrix

Description

Creates covariance matrices. Creates 3 types of matrices. 1) spatial correlation with no particular direction other than distance, 2) spatial correlation with one direction, 3) spatial correlation with 2 directions.

Usage

S.mat(dims, rhor=.95, rhoc=.95, method="SIM")

Arguments

dims

dimensions of the square matrix.

rhor

degree of autocorrelation in the row direction

rhoc

degree of autocorrelation in the column direction

method

one of the three methods available for creating correlation matrices for random effects related to spetial autocorrelation: 1) `SIM` spatial correlation with no particular direction other than distance,

2) `AR1` spatial correlation with one direction,

3) `AR2` spatial correlation with 2 directions.

Value

If everything is defined correctly the function returns:

$nn

the correlation matrix

Details

This function was created to stablish connections between levels of random effects such as row and columns in an experimental design. Hopefully users will find this helpful for adjusting for spatial correlation and avoid the use of correlation structures in the residual part of the model.

References

Covarrubias-Pazaran G (2016) Genome assisted prediction of quantitative traits using the R package sommer. PLoS ONE 11(6): doi:10.1371/journal.pone.0156744

See Also

The core functions of the package mmer and mmer2

Examples

R1 <- S.mat(10, method="SIM")
R2 <- S.mat(10, method="AR2")
layout(matrix(1:2,1,2))
image(R1);image(R2)