impute: Impute Conjugate Values

In eheinzen/gibbs.utils: Utilities for Gibbs Sampling

Impute Conjugate Values

Description

Impute Conjugate Values

Usage

impute_conj_mvnorm_mu(y, mu, impute, Q, mu0, tau0)

Arguments

y	A matrix of values drawn from the multivariate normal distribution whose mean we are to impute.
mu	the means we wish to impute. Should be the same size as y. No NA's are allowed; indices to impute are passed to impute. Instead, usually values are initialized and allowed to update within the MCMC framework.
impute	A logical matrix the size of y, indicating where to impute.
Q	the precision of the multivariate normal distribution from which y comes. Should be square, with dimension ncol(y).
mu0, tau0	The prior means/precisions for mu. Can be a matrix the same size as mu, in which case only the values corresponding to impute == TRUE are used, or a vector with length sum(impute).

eheinzen/gibbs.utils documentation built on Sept. 27, 2024, 9:03 p.m.