correlated.matrix: correlated.matrix

In synchrony: Methods for Computing Spatial, Temporal, and Spatiotemporal Statistics

Description

Create an ntimes x nspecies matrix with correlation rho, standard deviation sigma, and mean mu

Usage

correlated.matrix (rho = 0, sigma = 1, mu = 0, ntimes = 200, nspecies = 10)

Arguments

rho	Correlation between the columns of the matrix. This can be a single number describing the correlation between all columns, or the upper triangular portion of a correlation matrix describing the correlation between all pairs of columns. Default is 0
sigma	Standard deviation of the columns. Default is 1
mu	Mean of the columns. Default is 0
ntimes	Number of rows in the matrix. Default is 200
nspecies	Number of columns in the matrix. Default is 10

Details

This function is based on the Cholesky factorization method described by Legendre (2000).

Value

Returns a named list containing the following:

rho	Correlation(s) between the columns
sigma	Standard deviation of the columns
mu	Mean of the columns
community	ntimes x nspecies matrix

Author(s)

Tarik C. Gouhier ([e-mail])

References

Gouhier, T. C., F. Guichard, and A. Gonzalez. 2010. Synchrony and stability of food webs in metacommunities. The American Naturalist 175:E16-E34.

Legendre, P. 2000. Comparison of permutation methods for the partial correlation and partial mantel tests. Journal of Statistical Computation and Simulation 67:37-73.

Examples

mat=correlated.matrix(rho=0.85, sigma=30, mu=10, nspecies=10)
# Check sd of each column
apply(mat$community, 2, sd)
# Check mean of each column
apply(mat$community, 2, mean)
# Check correlation of matrix
community.sync(mat$community)

synchrony documentation built on March 26, 2020, 7:14 p.m.