Multivariate normal random numbers

Interface functions to use mvtnorm::rmvnorm() for hyperSpec-class() objects.

rmmvnorm(n, mean, sigma)

# S4 method for numeric,hyperSpec,matrix
rmmvnorm(n, mean, sigma)

# S4 method for numeric,hyperSpec,array
rmmvnorm(n, mean, sigma)

# S4 method for numeric,matrix,matrix
rmmvnorm(n, mean, sigma)

# S4 method for numeric,matrix,array
rmmvnorm(n, mean, sigma)

Arguments

n	vector giving the number of cases to generate for each group
mean	matrix with mean cases in rows
sigma	common covariance matrix or array (ncol(mean) x ncol(mean) x nrow(mean)) with individual covariance matrices for the groups.

Details

The mvtnorm method for hyperSpec objects supports producing multivariate normal data for groups with different mean but common covariance matrix, see the examples.

See also

mvtnorm::rmvnorm()

cov() and pooled.cov() about calculating covariance of hyperSpec objects.

Examples

## multiple groups, common covariance matrix

if (require("mvtnorm")) {
  pcov <- pooled.cov(faux_cell, faux_cell$region)
  rnd <- rmmvnorm(rep(10, 3), mean = pcov$mean, sigma = pcov$COV)
  plot(rnd, col = rnd$.group)
}
#> Loading required package: mvtnorm