Model averaged double robust estimation

Matthew Cefalu, Francesca Dominici, Nils Arvold, Giovanni Parmigiani

Research output: Contribution to journalArticle

Abstract

Researchers estimating causal effects are increasingly challenged with decisions on how to best control for a potentially high-dimensional set of confounders. Typically, a single propensity score model is chosen and used to adjust for confounding, while the uncertainty surrounding which covariates to include into the propensity score model is often ignored, and failure to include even one important confounder will results in bias. We propose a practical and generalizable approach that overcomes the limitations described above through the use of model averaging. We develop and evaluate this approach in the context of double robust estimation. More specifically, we introduce the model averaged double robust (MA-DR) estimators, which account for model uncertainty in both the propensity score and outcome model through the use of model averaging. The MA-DR estimators are defined as weighted averages of double robust estimators, where each double robust estimator corresponds to a specific choice of the outcome model and the propensity score model. The MA-DR estimators extend the desirable double robustness property by achieving consistency under the much weaker assumption that either the true propensity score model or the true outcome model be within a specified, possibly large, class of models. Using simulation studies, we also assessed small sample properties, and found that MA-DR estimators can reduce mean squared error substantially, particularly when the set of potential confounders is large relative to the sample size. We apply the methodology to estimate the average causal effect of temozolomide plus radiotherapy versus radiotherapy alone on one-year survival in a cohort of 1887 Medicare enrollees who were diagnosed with glioblastoma between June 2005 and December 2009.

Original languageEnglish (US)
JournalBiometrics
DOIs
StateAccepted/In press - 2016
Externally publishedYes

Fingerprint

Propensity Score
Robust Estimation
Robust Estimators
temozolomide
Uncertainty
Radiotherapy
Model
Model Averaging
Glioblastoma
Causal Effect
Medicare
Sample Size
radiotherapy
Research Personnel
Double Robustness
Confounding
Weighted Average
Model Uncertainty
model uncertainty
Mean Squared Error

Keywords

  • Causal inference
  • Confounding
  • Double robustness
  • Model averaging
  • Propensity score
  • Variable selection

ASJC Scopus subject areas

  • Statistics and Probability
  • Medicine(all)
  • Immunology and Microbiology(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)
  • Applied Mathematics

Cite this

Cefalu, M., Dominici, F., Arvold, N., & Parmigiani, G. (Accepted/In press). Model averaged double robust estimation. Biometrics. https://doi.org/10.1111/biom.12622

Model averaged double robust estimation. / Cefalu, Matthew; Dominici, Francesca; Arvold, Nils; Parmigiani, Giovanni.

In: Biometrics, 2016.

Research output: Contribution to journalArticle

Cefalu, M, Dominici, F, Arvold, N & Parmigiani, G 2016, 'Model averaged double robust estimation', Biometrics. https://doi.org/10.1111/biom.12622
Cefalu M, Dominici F, Arvold N, Parmigiani G. Model averaged double robust estimation. Biometrics. 2016. https://doi.org/10.1111/biom.12622
Cefalu, Matthew ; Dominici, Francesca ; Arvold, Nils ; Parmigiani, Giovanni. / Model averaged double robust estimation. In: Biometrics. 2016.
@article{f9cc9dcaed5e46ed9e5e2a03a3ddb6c5,
title = "Model averaged double robust estimation",
abstract = "Researchers estimating causal effects are increasingly challenged with decisions on how to best control for a potentially high-dimensional set of confounders. Typically, a single propensity score model is chosen and used to adjust for confounding, while the uncertainty surrounding which covariates to include into the propensity score model is often ignored, and failure to include even one important confounder will results in bias. We propose a practical and generalizable approach that overcomes the limitations described above through the use of model averaging. We develop and evaluate this approach in the context of double robust estimation. More specifically, we introduce the model averaged double robust (MA-DR) estimators, which account for model uncertainty in both the propensity score and outcome model through the use of model averaging. The MA-DR estimators are defined as weighted averages of double robust estimators, where each double robust estimator corresponds to a specific choice of the outcome model and the propensity score model. The MA-DR estimators extend the desirable double robustness property by achieving consistency under the much weaker assumption that either the true propensity score model or the true outcome model be within a specified, possibly large, class of models. Using simulation studies, we also assessed small sample properties, and found that MA-DR estimators can reduce mean squared error substantially, particularly when the set of potential confounders is large relative to the sample size. We apply the methodology to estimate the average causal effect of temozolomide plus radiotherapy versus radiotherapy alone on one-year survival in a cohort of 1887 Medicare enrollees who were diagnosed with glioblastoma between June 2005 and December 2009.",
keywords = "Causal inference, Confounding, Double robustness, Model averaging, Propensity score, Variable selection",
author = "Matthew Cefalu and Francesca Dominici and Nils Arvold and Giovanni Parmigiani",
year = "2016",
doi = "10.1111/biom.12622",
language = "English (US)",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",

}

TY - JOUR

T1 - Model averaged double robust estimation

AU - Cefalu, Matthew

AU - Dominici, Francesca

AU - Arvold, Nils

AU - Parmigiani, Giovanni

PY - 2016

Y1 - 2016

N2 - Researchers estimating causal effects are increasingly challenged with decisions on how to best control for a potentially high-dimensional set of confounders. Typically, a single propensity score model is chosen and used to adjust for confounding, while the uncertainty surrounding which covariates to include into the propensity score model is often ignored, and failure to include even one important confounder will results in bias. We propose a practical and generalizable approach that overcomes the limitations described above through the use of model averaging. We develop and evaluate this approach in the context of double robust estimation. More specifically, we introduce the model averaged double robust (MA-DR) estimators, which account for model uncertainty in both the propensity score and outcome model through the use of model averaging. The MA-DR estimators are defined as weighted averages of double robust estimators, where each double robust estimator corresponds to a specific choice of the outcome model and the propensity score model. The MA-DR estimators extend the desirable double robustness property by achieving consistency under the much weaker assumption that either the true propensity score model or the true outcome model be within a specified, possibly large, class of models. Using simulation studies, we also assessed small sample properties, and found that MA-DR estimators can reduce mean squared error substantially, particularly when the set of potential confounders is large relative to the sample size. We apply the methodology to estimate the average causal effect of temozolomide plus radiotherapy versus radiotherapy alone on one-year survival in a cohort of 1887 Medicare enrollees who were diagnosed with glioblastoma between June 2005 and December 2009.

AB - Researchers estimating causal effects are increasingly challenged with decisions on how to best control for a potentially high-dimensional set of confounders. Typically, a single propensity score model is chosen and used to adjust for confounding, while the uncertainty surrounding which covariates to include into the propensity score model is often ignored, and failure to include even one important confounder will results in bias. We propose a practical and generalizable approach that overcomes the limitations described above through the use of model averaging. We develop and evaluate this approach in the context of double robust estimation. More specifically, we introduce the model averaged double robust (MA-DR) estimators, which account for model uncertainty in both the propensity score and outcome model through the use of model averaging. The MA-DR estimators are defined as weighted averages of double robust estimators, where each double robust estimator corresponds to a specific choice of the outcome model and the propensity score model. The MA-DR estimators extend the desirable double robustness property by achieving consistency under the much weaker assumption that either the true propensity score model or the true outcome model be within a specified, possibly large, class of models. Using simulation studies, we also assessed small sample properties, and found that MA-DR estimators can reduce mean squared error substantially, particularly when the set of potential confounders is large relative to the sample size. We apply the methodology to estimate the average causal effect of temozolomide plus radiotherapy versus radiotherapy alone on one-year survival in a cohort of 1887 Medicare enrollees who were diagnosed with glioblastoma between June 2005 and December 2009.

KW - Causal inference

KW - Confounding

KW - Double robustness

KW - Model averaging

KW - Propensity score

KW - Variable selection

UR - http://www.scopus.com/inward/record.url?scp=85006152289&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006152289&partnerID=8YFLogxK

U2 - 10.1111/biom.12622

DO - 10.1111/biom.12622

M3 - Article

C2 - 27893927

AN - SCOPUS:85006152289

JO - Biometrics

JF - Biometrics

SN - 0006-341X

ER -