initParameters: Initialize the model parameters for the MCMC.

In NestedCategBayesImpute: Modeling, Imputing and Generating Synthetic Versions of Nested Categorical Data in the Presence of Impossible Combinations

Description

Initialize the model parameters for the MCMC.

Usage

initParameters(data, hyper, HHhead_at_group_level)

Arguments

data	A list object including all the necessary data variables needed by the sampler; output of the initData function.
hyper	Hyper parameters for the prior distributions.
HHhead_at_group_level	Logical indicator for data structure with respect to the household head. Set to TRUE if the household head has been moved to the household level and FALSE otherwise.

Value

A list of the initial values of the parameters.

alpha	Concentration parameter in the Dirichlet process for the group-level latent classes.
beta	Concentration parameter in the Dirichlet process for the individual-level latent classes. Currently, this is assumed to be the same within all group-level classes.
phi	Matrix of posterior samples for the individual-level probabilities for each individual-level variable by each pair of group-level and individual-level latent classes.
HHdata_all	The transposed household level data – each column represents each household.
lambda	A list of matrices of the group-level probabilities for each group-level variable by the group-level latent classes. Each matrix in the list is for each group-level variable.
u	Vector of the beta-distributed variables in the stick breaking representation of the group-level latent classes.
pi	Vector of the probabilities for the group-level latent classes.
v	Matrix of the beta-distributed variables in the stick breaking representation of the individual-level latent classes by the group-level latent classes.
omega	Matrix of the probabilities for the individual-level latent classes by the group-level latent classes.

Author(s)

Quanli Wang

NestedCategBayesImpute documentation built on May 2, 2019, 3:26 p.m.