rand.mnormal: Generate Random Samples from a Multivariate Normal...

In ldt: Automated Uncertainty Analysis

Generate Random Samples from a Multivariate Normal Distribution

Description

Use this function to get random samples from a multivariate normal distribution.

Usage

rand.mnormal(n, mu = NULL, sigma = NULL, p = NULL, byRow = TRUE)

Arguments

n	The number of samples to generate.
mu	The mean vector of the distribution. If NULL, it defaults to a zero vector of length p. If NA, it is set to a random vector.
sigma	The covariance matrix of the distribution. If NULL, it defaults to an identity matrix of size p x p. If NA, it is set to a random positive definite matrix.
p	The dimension of the distribution, if both mu and sigma are NA or NULL.
byRow	If TRUE, generated samples are stored in the rows. Otherwise, they are stored in the columns.

Value

A list containing the generated sample (p x n), mu, and sigma.

Examples

s1 <- rand.mnormal(10, mu = c(0, 0), sigma = matrix(c(1, 0.5, 0.5, 1), ncol = 2))
s2 <- rand.mnormal(10, mu = c(1,1), sigma = NA, p = 2)
s3 <- rand.mnormal(10, p = 2, byRow = FALSE) #standard normal

ldt documentation built on Sept. 11, 2024, 5:25 p.m.