On weighting approaches for missing data

Lingling Li, Changyu Shen, Xiaochun Li, James M. Robins

Research output: Contribution to journalArticle

33 Citations (Scopus)

Abstract

We review the class of inverse probability weighting (IPW) approaches for the analysis of missing data under various missing data patterns and mechanisms. The IPW methods rely on the intuitive idea of creating a pseudo-population of weighted copies of the complete cases to remove selection bias introduced by the missing data. However, different weighting approaches are required depending on the missing data pattern and mechanism. We begin with a uniform missing data pattern (i.e. a scalar missing indicator indicating whether or not the full data is observed) to motivate the approach. We then generalise to more complex settings. Our goal is to provide a conceptual overview of existing IPW approaches and illustrate the connections and differences among these approaches.

Original languageEnglish
Pages (from-to)14-30
Number of pages17
JournalStatistical Methods in Medical Research
Volume22
Issue number1
DOIs
StatePublished - Feb 2013

Fingerprint

Missing Data
Weighting
Inverse Probability Weighting
Selection Bias
Intuitive
Population
Scalar
Generalise

Keywords

  • inverse probability weighting
  • missing at random
  • missing data
  • missing not at random
  • monotone missing
  • non-monotone missing

ASJC Scopus subject areas

  • Epidemiology
  • Health Information Management
  • Statistics and Probability

Cite this

On weighting approaches for missing data. / Li, Lingling; Shen, Changyu; Li, Xiaochun; Robins, James M.

In: Statistical Methods in Medical Research, Vol. 22, No. 1, 02.2013, p. 14-30.

Research output: Contribution to journalArticle

Li, Lingling ; Shen, Changyu ; Li, Xiaochun ; Robins, James M. / On weighting approaches for missing data. In: Statistical Methods in Medical Research. 2013 ; Vol. 22, No. 1. pp. 14-30.
@article{6a0b9d9c7a474e20a1f823d475fb0c4f,
title = "On weighting approaches for missing data",
abstract = "We review the class of inverse probability weighting (IPW) approaches for the analysis of missing data under various missing data patterns and mechanisms. The IPW methods rely on the intuitive idea of creating a pseudo-population of weighted copies of the complete cases to remove selection bias introduced by the missing data. However, different weighting approaches are required depending on the missing data pattern and mechanism. We begin with a uniform missing data pattern (i.e. a scalar missing indicator indicating whether or not the full data is observed) to motivate the approach. We then generalise to more complex settings. Our goal is to provide a conceptual overview of existing IPW approaches and illustrate the connections and differences among these approaches.",
keywords = "inverse probability weighting, missing at random, missing data, missing not at random, monotone missing, non-monotone missing",
author = "Lingling Li and Changyu Shen and Xiaochun Li and Robins, {James M.}",
year = "2013",
month = "2",
doi = "10.1177/0962280211403597",
language = "English",
volume = "22",
pages = "14--30",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",
number = "1",

}

TY - JOUR

T1 - On weighting approaches for missing data

AU - Li, Lingling

AU - Shen, Changyu

AU - Li, Xiaochun

AU - Robins, James M.

PY - 2013/2

Y1 - 2013/2

N2 - We review the class of inverse probability weighting (IPW) approaches for the analysis of missing data under various missing data patterns and mechanisms. The IPW methods rely on the intuitive idea of creating a pseudo-population of weighted copies of the complete cases to remove selection bias introduced by the missing data. However, different weighting approaches are required depending on the missing data pattern and mechanism. We begin with a uniform missing data pattern (i.e. a scalar missing indicator indicating whether or not the full data is observed) to motivate the approach. We then generalise to more complex settings. Our goal is to provide a conceptual overview of existing IPW approaches and illustrate the connections and differences among these approaches.

AB - We review the class of inverse probability weighting (IPW) approaches for the analysis of missing data under various missing data patterns and mechanisms. The IPW methods rely on the intuitive idea of creating a pseudo-population of weighted copies of the complete cases to remove selection bias introduced by the missing data. However, different weighting approaches are required depending on the missing data pattern and mechanism. We begin with a uniform missing data pattern (i.e. a scalar missing indicator indicating whether or not the full data is observed) to motivate the approach. We then generalise to more complex settings. Our goal is to provide a conceptual overview of existing IPW approaches and illustrate the connections and differences among these approaches.

KW - inverse probability weighting

KW - missing at random

KW - missing data

KW - missing not at random

KW - monotone missing

KW - non-monotone missing

UR - http://www.scopus.com/inward/record.url?scp=84874501270&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874501270&partnerID=8YFLogxK

U2 - 10.1177/0962280211403597

DO - 10.1177/0962280211403597

M3 - Article

C2 - 21705435

AN - SCOPUS:84874501270

VL - 22

SP - 14

EP - 30

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

IS - 1

ER -