outlier_multi: Multivariate outliers

Description Usage Arguments Details Value References See Also Examples

Description

Find extreme values in ecological community data

Usage

1
2
outlier_multi(x, mult = 3, method = "bray", metric = "SD",
  plot = TRUE, perc = NULL, pick = NULL, ...)

Arguments

x

array of species data

mult

SD threshold for 'outliers' = the number of SD's away from the mean distance at which an outlier is defined; default is 3 times the SD

method

distance measure; follows vegdist

metric

defaults to 'SD'; only other alternative is 'MAD'

plot

logical, should a histogram of distances be drawn?

perc

numeric, returns outlier values beyond the specified percentile

pick

numeric, returns this number of most extreme outliers

...

further arguments passed to other methods

Details

Outlier analysis based on community distances. Emulates PC-ORD, but see also McCune and Grace (2002).

Value

A data frame with plot identifiers, calculated distances, and standard deviations from the mean distance, either with or without a histogram.

References

McCune, B., and J. B. Grace. 2002. Analysis of Ecological Communities. MjM Software, Gleneden Beach, Oregon, USA. 304 pp.

See Also

outlier_uni for univariate analog.

Other outlier functions: outlier_uni

Examples

1
2
3
4
5
6
7
8
require(vegan)
data(dune)
set.seed(19)
spe <- rbind(dune, runif(ncol(dune),0,1000)) # add an outlier row
outlier_multi(spe, mult=3, plot=FALSE)
outlier_multi(spe, mult=3)
outlier_multi(spe, perc=5)
outlier_multi(spe, pick=5)

phytomosaic/ecole documentation built on May 6, 2019, 3:45 p.m.