Non-Parametric Bayesian Multiple Imputation for Categorical Data
Estimating marginal and joint probabilities in imputed or synthetic da...
Create and initialize the Lcm model object
Use DPMPM models to impute missing data where there are no structural ...
Use DPMPM models to synthesize data where there are no structural zero...
Use DPMPM models to impute missing data where there are no structural ...
Fit GLM models for imputed or synthetic datasets
Convert imputed data to a dataframe, using the same setting from origi...
Convert disjointed structrual zeros to a dataframe, using the same set...
Perform MCMC diagnostics for kstar
Class "Rcpp_Lcm"
Plot estimated marginal probabilities from observed data vs imputed da...
Plot estimated marginal probabilities from observed data vs synthetic ...
Example dataframe for structrual zeros based on the NYMockexample data...
Bayesian Multiple Imputation for Large-Scale Categorical Data with Str...
Pool probability estimates from imputed or synthetic datasets
Pool estimates of fitted GLM models in imputed or synthetic datasets
Rcpp implemenation of the Lcm functions
Allow user to update the model with data matrix of same kind.
Example dataframe for input categorical data with missing values based...
These routines create multiple imputations of missing at random categorical data, and create multiply imputed synthesis of categorical data, with or without structural zeros. Imputations and syntheses are based on Dirichlet process mixtures of multinomial distributions, which is a non-parametric Bayesian modeling approach that allows for flexible joint modeling, described in Manrique-Vallier and Reiter (2014) <doi:10.1080/10618600.2013.844700>.