pool_estimated_probs

Pool probability estimates from imputed or synthetic datasets

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>.

Jingchen Hu

NPBayesImputeCat

Non-Parametric Bayesian Multiple Imputation for Categorical Data

pool_estimated_probs function

<dl> <dt>ComputeProbsResults</dt>
<dd>output from the compute_probs function</dd> <dt>method</dt>
<dd>choose between "imputation", "synthesis_full", "synthesis_partial"</dd></dl>

Arguments

Pool probability estimates from imputed or synthetic datasets — pool_estimated_probs

<dl>

 <dt>ComputeProbsResults</dt>
<dd>output from the compute_probs function</dd>

 <dt>method</dt>
<dd>choose between "imputation", "synthesis_full", "synthesis_partial"</dd>

</dl>

pool_estimated_probs: Pool probability estimates from imputed or synthetic datasets

Description

Usage

Value

Arguments