rMixtureNormalTransform: Generates data from a transformed mixture of multivariate...

In burrisk/midamix: Bayesian multiple imputation via mixtures

Description

Generates data from a transformed mixture of multivariate normal distributions.

Usage

rMixtureNormalTransform(n, mean = matrix(0, nrow = dim(sigma)[1], ncol =
  num_clusters), sigma = array(rep(diag(nrow(mean)), num_clusters), dim =
  c(nrow(mean), nrow(mean), num_clusters)), num_clusters = 1,
  prob = rep(1, num_clusters)/num_clusters,
  transformations = rep(list(function(y) y), nrow(mean)),
  method = c("eigen", "svd", "chol"))

Arguments

n	The number of samples to draw.
mean	A matrix corresponding to the means of each cluster.
sigma	An array with the covariance matrices of each cluster.
num_clusters	The number of clusters in the mixture.
prob	Probabilities of cluster membership.
transformations	A list of functions to apply to corresponding columns.
method	Method used for generating from a multivariate normal distribution.

Value

A matrix of samples from a transformed mixture of multivariate normal distributions.

burrisk/midamix documentation built on June 1, 2019, 12:49 p.m.