Adjustment uncertainty in effect estimation

Ciprian M Crainiceanu, Francesca Dominici, Giovanni Parmigiani

Research output: Contribution to journalArticle

Abstract

Often there is substantial uncertainty in the selection of confounders when estimating the association between an exposure and health. We define this type of uncertainty as 'adjustment uncertainty'. We propose a general statistical framework for handling adjustment uncertainty in exposure effect estimation for a large number of confounders, we describe a specific implementation, and we develop associated visualization tools. Theoretical results and simulation studies show that the proposed method provides consistent estimators of the exposure effect and its variance. We also show that, when the goal is to estimate an exposure effect accounting for adjustment uncertainty, Bayesian model averaging with posterior model probabilities approximated using information criteria can fail to estimate the exposure effect and can over- or underestimate its variance. We compare our approach to Bayesian model averaging using time series data on levels of fine particulate matter and mortality.

Original languageEnglish (US)
Pages (from-to)635-651
Number of pages17
JournalBiometrika
Volume95
Issue number3
DOIs
StatePublished - Sep 2008

Fingerprint

Uncertainty
Adjustment
uncertainty
Bayesian Model Averaging
Particulate Matter
Information Criterion
Bayes Theorem
Consistent Estimator
Probability Model
Time Series Data
Mortality
Estimate
Time series
time series analysis
particulates
Health
Visualization
Simulation Study
Bayesian model averaging
methodology

Keywords

  • Adjustment uncertainty
  • Air pollution
  • Bayesian model averaging

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Agricultural and Biological Sciences (miscellaneous)
  • Statistics and Probability
  • Mathematics(all)
  • Applied Mathematics

Cite this

Adjustment uncertainty in effect estimation. / Crainiceanu, Ciprian M; Dominici, Francesca; Parmigiani, Giovanni.

In: Biometrika, Vol. 95, No. 3, 09.2008, p. 635-651.

Research output: Contribution to journalArticle

Crainiceanu, CM, Dominici, F & Parmigiani, G 2008, 'Adjustment uncertainty in effect estimation', Biometrika, vol. 95, no. 3, pp. 635-651. https://doi.org/10.1093/biomet/asn015
Crainiceanu, Ciprian M ; Dominici, Francesca ; Parmigiani, Giovanni. / Adjustment uncertainty in effect estimation. In: Biometrika. 2008 ; Vol. 95, No. 3. pp. 635-651.
@article{f1f4a31c4ee44847ac8697ef9e8ca24c,
title = "Adjustment uncertainty in effect estimation",
abstract = "Often there is substantial uncertainty in the selection of confounders when estimating the association between an exposure and health. We define this type of uncertainty as 'adjustment uncertainty'. We propose a general statistical framework for handling adjustment uncertainty in exposure effect estimation for a large number of confounders, we describe a specific implementation, and we develop associated visualization tools. Theoretical results and simulation studies show that the proposed method provides consistent estimators of the exposure effect and its variance. We also show that, when the goal is to estimate an exposure effect accounting for adjustment uncertainty, Bayesian model averaging with posterior model probabilities approximated using information criteria can fail to estimate the exposure effect and can over- or underestimate its variance. We compare our approach to Bayesian model averaging using time series data on levels of fine particulate matter and mortality.",
keywords = "Adjustment uncertainty, Air pollution, Bayesian model averaging",
author = "Crainiceanu, {Ciprian M} and Francesca Dominici and Giovanni Parmigiani",
year = "2008",
month = "9",
doi = "10.1093/biomet/asn015",
language = "English (US)",
volume = "95",
pages = "635--651",
journal = "Biometrika",
issn = "0006-3444",
publisher = "Oxford University Press",
number = "3",

}

TY - JOUR

T1 - Adjustment uncertainty in effect estimation

AU - Crainiceanu, Ciprian M

AU - Dominici, Francesca

AU - Parmigiani, Giovanni

PY - 2008/9

Y1 - 2008/9

N2 - Often there is substantial uncertainty in the selection of confounders when estimating the association between an exposure and health. We define this type of uncertainty as 'adjustment uncertainty'. We propose a general statistical framework for handling adjustment uncertainty in exposure effect estimation for a large number of confounders, we describe a specific implementation, and we develop associated visualization tools. Theoretical results and simulation studies show that the proposed method provides consistent estimators of the exposure effect and its variance. We also show that, when the goal is to estimate an exposure effect accounting for adjustment uncertainty, Bayesian model averaging with posterior model probabilities approximated using information criteria can fail to estimate the exposure effect and can over- or underestimate its variance. We compare our approach to Bayesian model averaging using time series data on levels of fine particulate matter and mortality.

AB - Often there is substantial uncertainty in the selection of confounders when estimating the association between an exposure and health. We define this type of uncertainty as 'adjustment uncertainty'. We propose a general statistical framework for handling adjustment uncertainty in exposure effect estimation for a large number of confounders, we describe a specific implementation, and we develop associated visualization tools. Theoretical results and simulation studies show that the proposed method provides consistent estimators of the exposure effect and its variance. We also show that, when the goal is to estimate an exposure effect accounting for adjustment uncertainty, Bayesian model averaging with posterior model probabilities approximated using information criteria can fail to estimate the exposure effect and can over- or underestimate its variance. We compare our approach to Bayesian model averaging using time series data on levels of fine particulate matter and mortality.

KW - Adjustment uncertainty

KW - Air pollution

KW - Bayesian model averaging

UR - http://www.scopus.com/inward/record.url?scp=50949134553&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50949134553&partnerID=8YFLogxK

U2 - 10.1093/biomet/asn015

DO - 10.1093/biomet/asn015

M3 - Article

AN - SCOPUS:50949134553

VL - 95

SP - 635

EP - 651

JO - Biometrika

JF - Biometrika

SN - 0006-3444

IS - 3

ER -