Campus Units

Statistics

Document Type

Article

Publication Version

Accepted Manuscript

Publication Date

4-9-2014

Journal or Book Title

Biometrika

Volume

101

Issue

2

First Page

439

Last Page

448

DOI

10.1093/biomet/asu003

Abstract

Propensity score weighting adjustment is commonly used to handle unit nonresponse. When the response mechanism is nonignorable in the sense that the response probability depends directly on the study variable, a follow-up sample is commonly used to obtain an unbiased estimator using the framework of two-phase sampling, where the follow-up sample is assumed to respond completely. In practice, the follow-up sample is also subject to missingness. We consider propensity score weighting adjustment for nonignorable nonresponse when there are several follow-ups and the final follow-up sample is also subject to missingness. We propose a method-of-moments estimator for estimating parameters in the response probability. The proposed method can be implemented using the generalized method of moments and a consistent variance estimate can be obtained relatively easily. A limited simulation study shows the robustness of the proposed method. The proposed methods are applied to a Korean household survey of employment.

Comments

This is a pre-copyedited, author-produced PDF of an article accepted for publication in Biometrika following peer review. The version of record (Jae Kwang Kim, Jongho Im; Propensity score adjustment with several follow-ups. Biometrika 2014; 101 (2): 439-448) is available online at doi:10.1093/biomet/asu003. Posted with permission.

Copyright Owner

Biometrika Trust

Language

en

File Format

application/pdf

Published Version

Share

COinS