Propensity Score Analysis With Missing Data

Heining Cham, Stephen West

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Propensity score analysis is a method that equates treatment and control groups on a comprehensive set of measured confounders in observational (nonrandomized) studies. A successful propensity score analysis reduces bias in the estimate of the average treatment effect in a nonrandomized study, making the estimate more comparable with that obtained from a randomized experiment. This article reviews and discusses an important practical issue in propensity analysis, in which the baseline covariates (potential confounders) and the outcome have missing values (incompletely observed). We review the statistical theory of propensity score analysis and estimation methods for propensity scores with incompletely observed covariates. Traditional logistic regression and modern machine learning methods (e.g., random forests, generalized boosted modeling) as estimation methods for incompletely observed covariates are reviewed. Balance diagnostics and equating methods for incompletely observed covariates are briefly described. Using an empirical example, the propensity score estimation methods for incompletely observed covariates are illustrated and compared. (PsycINFO Database Record

Original languageEnglish (US)
JournalPsychological Methods
DOIs
StateAccepted/In press - Mar 10 2016

Fingerprint

Propensity Score
Observational Studies
Logistic Models
Control Groups

Keywords

  • Machine learning
  • Missing data
  • Nonrandomization
  • Propensity score

ASJC Scopus subject areas

  • Psychology (miscellaneous)

Cite this

Propensity Score Analysis With Missing Data. / Cham, Heining; West, Stephen.

In: Psychological Methods, 10.03.2016.

Research output: Contribution to journalArticle

@article{2d56003883b34bd98f8230565b399ee1,
title = "Propensity Score Analysis With Missing Data",
abstract = "Propensity score analysis is a method that equates treatment and control groups on a comprehensive set of measured confounders in observational (nonrandomized) studies. A successful propensity score analysis reduces bias in the estimate of the average treatment effect in a nonrandomized study, making the estimate more comparable with that obtained from a randomized experiment. This article reviews and discusses an important practical issue in propensity analysis, in which the baseline covariates (potential confounders) and the outcome have missing values (incompletely observed). We review the statistical theory of propensity score analysis and estimation methods for propensity scores with incompletely observed covariates. Traditional logistic regression and modern machine learning methods (e.g., random forests, generalized boosted modeling) as estimation methods for incompletely observed covariates are reviewed. Balance diagnostics and equating methods for incompletely observed covariates are briefly described. Using an empirical example, the propensity score estimation methods for incompletely observed covariates are illustrated and compared. (PsycINFO Database Record",
keywords = "Machine learning, Missing data, Nonrandomization, Propensity score",
author = "Heining Cham and Stephen West",
year = "2016",
month = "3",
day = "10",
doi = "10.1037/met0000076",
language = "English (US)",
journal = "Psychological Methods",
issn = "1082-989X",
publisher = "American Psychological Association Inc.",

}

TY - JOUR

T1 - Propensity Score Analysis With Missing Data

AU - Cham, Heining

AU - West, Stephen

PY - 2016/3/10

Y1 - 2016/3/10

N2 - Propensity score analysis is a method that equates treatment and control groups on a comprehensive set of measured confounders in observational (nonrandomized) studies. A successful propensity score analysis reduces bias in the estimate of the average treatment effect in a nonrandomized study, making the estimate more comparable with that obtained from a randomized experiment. This article reviews and discusses an important practical issue in propensity analysis, in which the baseline covariates (potential confounders) and the outcome have missing values (incompletely observed). We review the statistical theory of propensity score analysis and estimation methods for propensity scores with incompletely observed covariates. Traditional logistic regression and modern machine learning methods (e.g., random forests, generalized boosted modeling) as estimation methods for incompletely observed covariates are reviewed. Balance diagnostics and equating methods for incompletely observed covariates are briefly described. Using an empirical example, the propensity score estimation methods for incompletely observed covariates are illustrated and compared. (PsycINFO Database Record

AB - Propensity score analysis is a method that equates treatment and control groups on a comprehensive set of measured confounders in observational (nonrandomized) studies. A successful propensity score analysis reduces bias in the estimate of the average treatment effect in a nonrandomized study, making the estimate more comparable with that obtained from a randomized experiment. This article reviews and discusses an important practical issue in propensity analysis, in which the baseline covariates (potential confounders) and the outcome have missing values (incompletely observed). We review the statistical theory of propensity score analysis and estimation methods for propensity scores with incompletely observed covariates. Traditional logistic regression and modern machine learning methods (e.g., random forests, generalized boosted modeling) as estimation methods for incompletely observed covariates are reviewed. Balance diagnostics and equating methods for incompletely observed covariates are briefly described. Using an empirical example, the propensity score estimation methods for incompletely observed covariates are illustrated and compared. (PsycINFO Database Record

KW - Machine learning

KW - Missing data

KW - Nonrandomization

KW - Propensity score

UR - http://www.scopus.com/inward/record.url?scp=84959865972&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959865972&partnerID=8YFLogxK

U2 - 10.1037/met0000076

DO - 10.1037/met0000076

M3 - Article

C2 - 26962757

AN - SCOPUS:84959865972

JO - Psychological Methods

JF - Psychological Methods

SN - 1082-989X

ER -