par.nmds: A parallelized function for estimating species diversity
In parfossil: Parallelized functions for palaeoecological and palaeogeographical analysis

Description Usage Arguments Details Value Note Author(s) References See Also Examples

Estimate the diversity of a sample(s) using a number of species diversity estimators.

1	par.nmds(dmat, mindim = 1, maxdim = 2, nits = 10, iconf = 0, epsilon = 1e-12, maxit = 500, trace = FALSE)

`dmat`	Lower triangle distance matrix
`mindim`	optional, the minmum number of dimensions to use for an analysis; default is 1
`maxdim`	optional, the maximum number of dimensions to use for an analysis; default is 2
`nits`	optional, the number of iterations; how many times the data should be initially placed at random; default is 10
`iconf`	optional, initial configuration. If not specified, then a random configuration is used.
`epsilon`	optional, acceptable difference in stress.
`maxit`	optional, maximum number of iterations.
`trace`	if TRUE, will write progress indicator to the screen.

Non-Metric Multidimensional Scaling (NMDS) is designed to find an optimal arrangement for a set of points in a reduced dimensional space.

conf : list of configurations.

stress : list of final stress values.

r2 : total variance explained by each configuration. The first results are for the lowest number of dimensions (total number is (mindim - maxdim + 1) * nits).

This is slight modification of the nmds function found in the ecodist package, that has been changed to allow for parallelization of runs. This version of the function uses the foreach() function to parallelize the resampling loop, so any backend that can be used with that package can be used to enable the parallel processing with this package.

Sarah Goslee with modifications from Matthew Vavrek

The original nmds function used as the basis for this parallelized version comes from the ecodist package.

Goslee, S.C., Urban, D.L. 2007. The ecodist Package for Dissimilarity-based Analysis of Ecological Data. Journal of Statistical Software. 22(7):1-19.

ecol.dist

## Not run: 
#comparison of run times between the serial and parallel versions on the estimator
#please note that this example is designed for a multicore OS X or Linux computer
library(doMC)
registerDoMC()
data(fdata.mat)
par.nmds(ecol.dist(fdata.mat))

## End(Not run)