Using data augmentation to correct for nonignorable nonresponse when surrogate data are available: An application to the distribution of hourly pay
Using data augmentation to correct for nonignorable nonresponse when surrogate data are available: An application to the distribution of hourly pay
This paper develops a data augmentation method to estimate the distribution function of a variable, which is partially observed, under a nonignorable missing data mechanism, and where surrogate data are available. An application to the estimation of hourly pay distributions using UK Labour Force Survey (LFS) data provides the main motivation.
In addition to considering a standard parametric data augmentation method, we consider the use of hot deck imputation methods as part of the data augmentation procedure to improve the robustness of the method. The proposed method is compared with standard methods based upon an ignorable missing data mechanism, both in a simulation study and in the LFS application. The focus is on reducing bias in point estimation, but variance estimation using multiple imputation is also considered briefly.
distribution function estimation, imputation, measurement error, missing data, multiple imputation, rejection sampling
Southampton Statistical Sciences Research Institute, University of Southampton
Beissel-Durrant, Gabriele
9630d22e-5f26-4407-bcfd-9674a03b4ee1
Skinner, Chris
dec5ef40-49ef-492a-8a1d-eb8c6315b8ce
20 September 2004
Beissel-Durrant, Gabriele
9630d22e-5f26-4407-bcfd-9674a03b4ee1
Skinner, Chris
dec5ef40-49ef-492a-8a1d-eb8c6315b8ce
Beissel-Durrant, Gabriele and Skinner, Chris
(2004)
Using data augmentation to correct for nonignorable nonresponse when surrogate data are available: An application to the distribution of hourly pay
(S3RI Methodology Working Papers, M04/10)
Southampton, UK.
Southampton Statistical Sciences Research Institute, University of Southampton
30pp.
Record type:
Monograph
(Working Paper)
Abstract
This paper develops a data augmentation method to estimate the distribution function of a variable, which is partially observed, under a nonignorable missing data mechanism, and where surrogate data are available. An application to the estimation of hourly pay distributions using UK Labour Force Survey (LFS) data provides the main motivation.
In addition to considering a standard parametric data augmentation method, we consider the use of hot deck imputation methods as part of the data augmentation procedure to improve the robustness of the method. The proposed method is compared with standard methods based upon an ignorable missing data mechanism, both in a simulation study and in the LFS application. The focus is on reducing bias in point estimation, but variance estimation using multiple imputation is also considered briefly.
Text
9189-01.pdf
- Author's Original
More information
Submitted date: 20 September 2004
Published date: 20 September 2004
Keywords:
distribution function estimation, imputation, measurement error, missing data, multiple imputation, rejection sampling
Identifiers
Local EPrints ID: 9189
URI: http://eprints.soton.ac.uk/id/eprint/9189
PURE UUID: 818b4300-5b80-4162-8c0e-fa5e5188f8e1
Catalogue record
Date deposited: 28 Jun 2006
Last modified: 15 Mar 2024 04:55
Export record
Contributors
Author:
Gabriele Beissel-Durrant
Author:
Chris Skinner
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics