Multiple imputation by chained equations: what is it and how does it work?

International Journal of Methods in Psychiatric Research - Tập 20 Số 1 - Trang 40-49 - 2011
Melissa Azur1, Elizabeth A. Stuart1, Constantine Frangakis2, Philip J. Leaf1
1Department of Mental Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
2Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MA, USA

Tóm tắt

AbstractMultivariate imputation by chained equations (MICE) has emerged as a principled method of dealing with missing data. Despite properties that make MICE particularly useful for large imputation procedures and advances in software development that now make it accessible to many researchers, many psychiatric researchers have not been trained in these methods and few practical resources exist to guide researchers in the implementation of this technique. This paper provides an introduction to the MICE method with a focus on practical aspects and challenges in using this method. A brief review of software programs available to implement MICE and then analyze multiply imputed data is also provided. Copyright © 2011 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

10.1111/j.1467‐9876.2007.00613.x

10.1177/0962280206075310

BrandJ.P.L.(1999)Development implementation and evaluation of multiple imputation strategies for the statistical analysis of incomplete data sets unpublished Erasmus University Rotterdam.

10.1037/1082‐989X.6.4.330

10.1002/sim.3239

GelmanA. HillJ. YajimaM. SuY. PittauM.(2011)mi: Missing Data Imputation and Model Checking. Package for the R statistical software.http://lib.stat.cmu.edu/R/CRAN/[27 January 2011]

10.1207/S15328007SEM1001_4

10.1146/annurev.psych.58.110405.085530

10.1007/s11121‐007‐0070‐9

10.1093/oxfordjournals.aje.a117592

10.1002/sim.2787

10.1177/0962280208101273

10.1198/000313007X172556

Jacobusse G., 2005, WinMICE User's Manual for WinMICE Prototype Version 0.1

10.1093/aje/kwp425

10.1002/sim.2343

LumleyT.(2008)Mitools: tools for multiple imputation of missing data. Package for the R statistical software package.http://cran.r‐project.org/web/packages/mitools/[14 May 2008].

10.1207/S15326918CS0501_2

10.1016/j.jclinepi.2006.01.009

Muthen L.K., Mplus User's Guide

10.1002/sim.3731

R Development Core Team, 2008, R: A Language and Environment for Statistical Computing

Raghunathan T.W., 2001, A multivariate technique for multiply imputing missing values using a sequence of regression models, Survey Methodology, 27, 85

Raghunathan T.E., 2002, IVEware: Imputation and Variance Estimation Software User Guide

10.1177/1536867X0500500204

10.1177/1536867X0900900205

SAS Institute Inc., 2008, SAS/STAT User's Guide 9.2

10.1177/096228029900800102

10.1111/1467-9574.00218

10.1111/1467‐9574.00218

10.1198/016214505000001375

Scientific Software International, 2006, Lisrel 8.8

SPSS Inc., 2009, Amos 18.0. SPSS Missing Values 17.0

SPSS Inc., 2009, SPSS Missing Values 17.0

StataCorp, 2009, Stata Multiple‐imputation Reference Manual

10.1093/aje/kwp026

Su Y‐S., Multiple imputation with diagnostics (mi) in R: Opening windows into the black box, Journal of Statistical Software

10.1177/0962280206074463

Van Buuren S., MICE: Multivariate Imputation by Chained Equations in R, Journal of Statistical Software

10.1177/0962280206074464

Yuan Y.C., 2000, Proceedings of the Twenty‐Fifth Annual SAS Users Group International Conference

10.1098/rsta.2008.0038

Thông tin
Thông tin xuất bản

International Journal of Methods in Psychiatric Research

Tập 20 Số 1

40-49

Thông tin tác giả