simulate_rel_sd: Simulate measurement variability

In ssokolen/metcourse: Tools for Metabolic Time-course Data

Description

Simulates relative standard deviations (coefficients of variation) using a mixture of 2 normal distributions (one subpopulation of less variable measurements and on subpopulation of more variable ones).

Usage

simulate_rel_sd(n, par1, par2 = NULL, p = 0.05, con = c(0, 1))

Arguments

n	Number of samples to draw.
par1	Parameters (mean, sd) for first distribution.
par2	Optional parameters (mean, sd) for second distribution.
p	Proportion of samples to draw using par1 vs. par2. Note, p is ignored if par2 is not specified.
con	Optional vector of lower and upper constraints on generated samples. By default, relative standard deviations are limited to between 0 and 1.

Value

A vector of relative standard deviation values.

Examples

out <- simulate_rel_sd(10000, c(.04, .02), c(0.11, 0.02), 0.7, c(0, 0.20))
hist(out, 20, probability = TRUE,
     main = '', xlab = 'Relative standard deviation')

ssokolen/metcourse documentation built on May 30, 2019, 8:43 a.m.