Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/271
Title: Local multiple imputation
Authors: AERTS, Marc 
CLAESKENS, Gerda 
HENS, Niel 
MOLENBERGHS, Geert 
Issue Date: 2002
Publisher: OXFORD UNIV PRESS
Source: Biometrika, 89(2). p. 375-388
Abstract: Dealing with missing data via parametric multiple imputation methods usually implies stating several strong assumptions both about the distribution of the data and about underlying regression relationships.If such parametric assumptions do not hold, the multiply imputed data are not appropriate and might produce inconsistent estimators and thus misleading results. In this paper, a fully nonparametric and a semiparametric imputation method are studied, both based on local resampling principles. It is shown that the final estimator, based on these local imputations, is consistent under fewer or no parametric assumptions. Asymptotic expressions for bias, variance and mean squared error are derived, showing the theoretical impact of the different smoothing parameters. Simulations illustrate the usefulness and applicability of the method.
Keywords: bootstrap; Kernel weight; missing value; multiple imputation; nonparametric imputation; nonresponse; semiparametric imputation
Document URI: http://hdl.handle.net/1942/271
ISSN: 0006-3444
e-ISSN: 1464-3510
DOI: 10.1093/biomet/89.2.375
ISI #: 000176520500009
Rights: Copyright Biometrika Trust 2002
Category: A1
Type: Journal Contribution
Validations: ecoom 2003
Appears in Collections:Research publications

Files in This Item:
File Description SizeFormat 
a.pdf
  Restricted Access
Published version134.52 kBAdobe PDFView/Open    Request a copy
a.pdfPeer-reviewed author version135.51 kBAdobe PDFView/Open
Show full item record

SCOPUSTM   
Citations

26
checked on Sep 2, 2020

WEB OF SCIENCETM
Citations

32
checked on Dec 4, 2022

Page view(s)

78
checked on Sep 7, 2022

Download(s)

206
checked on Sep 7, 2022

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.