noise: Noise

In phytomosaic/vuln: Niche-based Vulnerability of Species and Communities

Description

Add random uncertainty for sensitivity analysis.

Usage

noise(z, perc = 1e-04, unif = TRUE, ...)

Arguments

z	vector, matrix, or data.frame of observations.
perc	numeric, percent of data range.
unif	logical, use default uniform distribution? Alternative is to use normal distribution.
...	further arguments passed to other functions.

Details

Adds noise, ignoring NA values. If unif=TRUE, then uniformly-distributed noise is bounded within perc percent of the original data range. If unif=FALSE, then normally-distributed noise has standard deviation equal to perc percent of the original data range. Useful for perturbing values for sensitivity analysis.

Value

Matrix of noisified data of same dimensions as 'z'.

Examples

# iris data
x  <- iris[,1:3]
xn <- noise(x, perc=1)
pairs(x, lower.panel=NULL)
pairs(xn, lower.panel=NULL)

phytomosaic/vuln documentation built on Sept. 21, 2019, 8:23 a.m.