MixtureModel-class: An object for running MCMC simulations.

Description Slots

Description

BatchModel and MarginalModel both inherit from this class.

Slots

k

An integer value specifying the number of latent classes.

hyperparams

An object of class 'Hyperparameters' used to specify the hyperparameters of the model.

theta

the means of each component and batch

sigma2

the variances of each component and batch

nu.0

the shape parameter for sigma2

sigma2.0

the rate parameter for sigma2

pi

mixture probabilities which are assumed to be the same for all batches

mu

overall mean

tau2

overall variance

data

the data for the simulation.

data.mean

the empirical means of the components

data.prec

the empirical precisions

z

latent variables

zfreq

table of latent variables

probz

n x k matrix of probabilities

u

chi-square draws for controlling t-distribution

logprior

log likelihood of prior: log(p(sigma2.0)p(nu.0)p(mu))

loglik

log likelihood: ∑ p_k Φ(θ_k, σ_k)

mcmc.chains

an object of class 'McmcChains' to store MCMC samples

batch

an integer-vector numbering the different batches. Must the same length as data.

batchElements

a vector labeling from which batch each observation came from

modes

the values of parameters from the iteration which maximizes log likelihood and log prior

mcmc.params

An object of class 'McmcParams'

label_switch

length-one logical indicating problems with label switching

marginal_lik

the marginal likelihood of the model

.internal.constraint

Constraint on parameters. For internal use only.

.internal.counter

For internal use only.

marginal_lik

scalar for marginal likelihood


scristia/CNPBayes documentation built on Aug. 9, 2020, 7:31 p.m.