rnddist {labdsv} | R Documentation |
Random Distance
Description
Calculates a random distance matrix for use in null model analysis.
Usage
rnddist(size, method='metric', sat = 1.0, upper=FALSE,
diag=FALSE)
Arguments
size |
the number of items to calculate the distances for |
method |
the desired properties of the matrix. Must be either ‘metric’ or ‘euclidean’ |
sat |
a saturation coefficient to set an upper limit less than 1.0 that truncates maximum values to simulate a dissimilarity rather than a distance |
upper |
logical: whether to print the upper triangle (default=FALSE) |
diag |
logical: whether to print the diagonal (default=FALSE) |
Details
Generates a matrix of size^2
uniform random numbers
and passes the matrix to metrify
or
euclidify
to ensure the metric or
euclidean properties of the distances. Values are normalized
to a maximum of 1.0.
Value
A dissimilarity object of class ‘dist’
Author(s)
David W. Roberts droberts@montana.edu
See Also
Examples
x <- rnddist(100)
pco.x <- pco(x)
plot(pco.x)
[Package labdsv version 2.1-0 Index]