orank {dave}

R Documentation

Ranking by orthogonal components (RANK)

Description

Given a correlation matrix of rows or columns this selects the variable sharing a maximum variance with all others and declares this rank 1. Reduces the matrix (covariances, correlations) by the contribution of the variable ranked first. Repeats the process to derive consecutive ranks until no variance is left.

Usage

orank(veg, use, rlimit=5, y=1, x.axis=NULL, y.axis=NULL,...)
orank1(veg, use, rlimit=5, y=1, x.axis=NULL, y.axis=NULL)

## Default S3 method:
orank(veg, use, rlimit=5, y=1, x.axis=NULL, y.axis=NULL,...)
## S3 method for class 'orank'
plot(x,...)
## S3 method for class 'orank'
summary(object,...)

Arguments

veg	This is a vegetation data frame, releves are rows, species columns
use	Either "rows" or "columns"
rlimit	The maximum number of ranks to be determined
y	Transformation of species scores: x'= x exp(y)
x.axis	Horizontal axis used for plotting result in a sampling plan
y.axis	Vertical axis used for plotting result in a sampling plan
x	An object of class "orank"
...	Further variables used for printing
object	An object of class "orank"

Details

If x-axis=NULL or y-axis=NULL then a pcoa-ordination is computed and the first two axes used for plotting

Value

An object of class "orank" with at least the following items:

use	Either "rows" or "columns"
n.ranks	The number of ranks
var.names	Names of the ranked variables
var.explained	Explained variance of the ranked variables
var.percent	Percentage of the variance explained
cum.var	Cummulative variance of ranked variables, percentage
x.axis	The same as input parameter x.axis
y.axis	The same as input parameter y.axis

Note

The present function exclusively relies on a correlation matrix, function cor.

Author(s)

Otto Wildi

References

Orloci, L. 1973. Ranking characters by a dispersion criterion. Nature 244: 371–373.

Wildi, O. 2017. Data Analysis in Vegetation Ecology. 3rd ed. CABI, Oxfordshire, Boston.

Examples

# Uses vegetation data frame sveg with vegetation data
# and ssit with corresponding x- and y-axes scores
x.axis=ssit$x.axis ; y.axis=ssit$y.axis
o.orank<- orank(sveg,use="rows",rlimit=5,y=0.25,x.axis,y.axis)
plot(o.orank)

---