NMDS: Nonmetric Multidimensional Scaling

In barebone: Multivariate Methods in Ecology with R: Basic Algorithms

Description

Nonmetric multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis (Kruskal 1964)

Usage

NMDS(D, k = 2, u, method = "BFGS", ...)

Arguments

D	Dissimilarity structure.
k	Number of dimensions.
u	Optional starting configuration. A random start is used if missing.
method	Minimization method for optim.
...	Other arguments passed to optim.

Details

Function consists of stress function, derivative function of the stress, and call to optim.

Value

Function returns a result object of optim. For most items, see the documentation of optim. The following items have a specific meaning in NMDS:

par	The final configuration.
value	The final stress.

Author(s)

Jari Oksanen

References

Kruskal, J. B. (1964) Nonmetric multidimensional scaling: a numerical method. Psychometrika 29, 115–129.

See Also

isoMDS function (MASS package) implements the same algorithm but written in C. See optim for the optimization routine and isoreg for isotonic regression.

barebone documentation built on May 2, 2019, 5:17 p.m.