A toolkit in SAS for the evaluation of multiple imputation methods

Jaap P.L. Brand; Stef van Buuren; Karin Groothuis-Oudshoorn; Edzard S. Gelsema

doi:10.1111/1467-9574.00219

A toolkit in SAS for the evaluation of multiple imputation methods

Jaap P.L. Brand, Stef van Buuren, Karin Groothuis-Oudshoorn, Edzard S. Gelsema

Research output: Contribution to journal › Article › Academic › peer-review

32 Citations (Scopus)

Abstract

This paper outlines a strategy to validate multiple imputation methods. Rubin's criteria for proper multiple imputation are the point of departure. We describe a simulation method that yields insight into various aspects of bias and efficiency of the imputation process. We propose a new method for creating incomplete data under a general Missing At Random (MAR) mechanism. Software implementing the validation strategy is available as a SAS/IML module. The method is applied to investigate the behavior of polytomous regression imputation for categorical data.

Original language	English
Pages (from-to)	36-45
Number of pages	10
Journal	Statistica Neerlandica
Volume	57
Issue number	1
DOIs	https://doi.org/10.1111/1467-9574.00219
Publication status	Published - Feb 2003
Externally published	Yes

Keywords

multiple imputation
proper imputation
missing data mechanism
simulation

Access to Document

10.1111/1467-9574.00219

Cite this

@article{3d77592d313645749696cc9349d1fc1c,

title = "A toolkit in SAS for the evaluation of multiple imputation methods",

abstract = "This paper outlines a strategy to validate multiple imputation methods. Rubin's criteria for proper multiple imputation are the point of departure. We describe a simulation method that yields insight into various aspects of bias and efficiency of the imputation process. We propose a new method for creating incomplete data under a general Missing At Random (MAR) mechanism. Software implementing the validation strategy is available as a SAS/IML module. The method is applied to investigate the behavior of polytomous regression imputation for categorical data.",

keywords = "multiple imputation, proper imputation, missing data mechanism, simulation",

author = "Brand, {Jaap P.L.} and {van Buuren}, Stef and Karin Groothuis-Oudshoorn and Gelsema, {Edzard S.}",

year = "2003",

month = feb,

doi = "10.1111/1467-9574.00219",

language = "English",

volume = "57",

pages = "36--45",

journal = "Statistica Neerlandica",

issn = "0039-0402",

publisher = "Wiley-Blackwell",

number = "1",

}

TY - JOUR

T1 - A toolkit in SAS for the evaluation of multiple imputation methods

AU - Brand, Jaap P.L.

AU - van Buuren, Stef

AU - Groothuis-Oudshoorn, Karin

AU - Gelsema, Edzard S.

PY - 2003/2

Y1 - 2003/2

N2 - This paper outlines a strategy to validate multiple imputation methods. Rubin's criteria for proper multiple imputation are the point of departure. We describe a simulation method that yields insight into various aspects of bias and efficiency of the imputation process. We propose a new method for creating incomplete data under a general Missing At Random (MAR) mechanism. Software implementing the validation strategy is available as a SAS/IML module. The method is applied to investigate the behavior of polytomous regression imputation for categorical data.

AB - This paper outlines a strategy to validate multiple imputation methods. Rubin's criteria for proper multiple imputation are the point of departure. We describe a simulation method that yields insight into various aspects of bias and efficiency of the imputation process. We propose a new method for creating incomplete data under a general Missing At Random (MAR) mechanism. Software implementing the validation strategy is available as a SAS/IML module. The method is applied to investigate the behavior of polytomous regression imputation for categorical data.

KW - multiple imputation

KW - proper imputation

KW - missing data mechanism

KW - simulation

U2 - 10.1111/1467-9574.00219

DO - 10.1111/1467-9574.00219

M3 - Article

VL - 57

SP - 36

EP - 45

JO - Statistica Neerlandica

JF - Statistica Neerlandica

SN - 0039-0402

IS - 1

ER -

A toolkit in SAS for the evaluation of multiple imputation methods

Abstract

Keywords

Access to Document

Fingerprint

Cite this