fcondYXA: Conditional posterior distribution of the latents Y

In konkam/BNPdensity: Ferguson-Klass Type Algorithm for Posterior Normalized Random Measures

Conditional posterior distribution of the latents Y

Description

This function simulates form the conditional posterior distribution of the latents Y.

Usage

fcondYXA(x, distr, Tau, J, sigma)

Details

For internal use.

Examples

## The function is currently defined as
function(x, distr = 1, Tau, J, sigma) {
  K <- matrix(NA, nrow = length(Tau), ncol = length(x))
  for (i in seq(Tau)) {
    K[i, ] <- dk(x, distr = distr, mu = Tau[i], sigma = sigma) *
      J[i]
  }
  pK <- prop.table(K, margin = 2)
  y <- apply(pK, 2, function(x) sample(Tau, size = 1, prob = x))
  return(y)
}

konkam/BNPdensity documentation built on March 14, 2024, 7:15 a.m.