Publication Cover

The American Statistician

American Statistician, The
Volume 71, 2017 - Issue 2
1,930
Views
23
CrossRef citations to date
0
Altmetric
Statistical Computing and Graphics

An Empirical Comparison of Multiple Imputation Methods for Categorical Data

ORCID Icon, &
Pages 162-170
Received 01 Aug 2015
Accepted author version posted online: 19 Jan 2017
Published online:06 Jul 2017
 

ABSTRACT

Multiple imputation is a common approach for dealing with missing values in statistical databases. The imputer fills in missing values with draws from predictive models estimated from the observed data, resulting in multiple, completed versions of the database. Researchers have developed a variety of default routines to implement multiple imputation; however, there has been limited research comparing the performance of these methods, particularly for categorical data. We use simulation studies to compare repeated sampling properties of three default multiple imputation methods for categorical data, including chained equations using generalized linear models, chained equations using classification and regression trees, and a fully Bayesian joint distribution based on Dirichlet process mixture models. We base the simulations on categorical data from the American Community Survey. In the circumstances of this study, the results suggest that default chained equations approaches based on generalized linear models are dominated by the default regression tree and Bayesian mixture model approaches. They also suggest competing advantages for the regression tree and Bayesian mixture model approaches, making both reasonable default engines for multiple imputation of categorical data. Supplementary material for this article is available online.

Funding

This research was supported by grants from the National Science Foundation (SES-11-31897) and the Alfred P. Sloan Foundation (G-2015-20166003).

Log in via your institution

Log in to Taylor & Francis Online

PDF download + Online access

  • 48 hours access to article PDF & online version
  • Article PDF can be downloaded
  • Article PDF can be printed
EUR 47.00 Add to cart

Issue Purchase

  • 30 days online access to complete issue
  • Article PDFs can be downloaded
  • Article PDFs can be printed
EUR 74.00 Add to cart

Purchase access via tokens

  • Choose from packages of 10, 20, and 30 tokens
  • Can use on articles across multiple libraries & subject collections
  • Article PDFs can be downloaded & printed
From EUR 400.00
per package
Learn more
* Local tax will be added as applicable
 

Related research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.