Modeling, Imputing and Generating Synthetic Versions of Nested Categorical Data in the Presence of Impossible Combinations
Checking a data matrix of households for the possible/impossible statu...
Checking a data matrix of households for the possible/impossible statu...
The new implementation of checkconstraints and will evently replace ch...
Michael: Edit here
Generate the desired number of impossible households required to obser...
Generate 2D count table for two integer-valued vectors.
Generate histogram count for an integer-valued vector.
Convert a household data matrix to the corresponding individual member...
Initialize the input data structure.
Initilize the misising data structure from input data
Set the output structure for saving posterior samples of parameters.
Initialize the model parameters for the MCMC.
Run the mcmc sampler for the model.
Update household (group) level latent class indexes.
Rcpp implementation for sampling household data without constraints.
Update individual level latent class indexes.
Sample and update missing data
Update alpha.
Update beta.
Update lambda.
Update lambda.
Update omega and v.
Update omega and v.
Update phi.
Update phi.
Update pi and u.
Update pi and u.
This tool set provides a set of functions to fit the nested Dirichlet process mixture of products of multinomial distributions (NDPMPM) model for nested categorical household data in the presence of impossible combinations. It has direct applications in imputing missing values for and generating synthetic versions of nested household data.