logacceptance_ratio_logu: Metropolis-Hastings ratio for the conditional of logU

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

Metropolis-Hastings ratio for the conditional of logU

Description

This function computes the Metropolis-Hastings ratio to decide whether to accept or reject a new value for logU.

Usage

logacceptance_ratio_logu(logu, logu_prime, n, r, gamma, kappa, a, delta)

Arguments

logu	Real, log of the latent variable U at the current iteration.
logu_prime	Real, log of the new proposed latent variable U.
n	Integer, number of data points.
r	Integer, number of clusters.
kappa	Positive real. A parameter of the NRMI process.
a	Positive real. Total mass of the centering measure.
delta	Scale of the Metropolis-Hastings proposal distribution

BNPdensity documentation built on April 1, 2023, 12:10 a.m.