Doubly robust estimation combines two adjustment strategies:
An outcome model (regression of \(Y\) on \(X\) and treatment status)
A treatment model (propensity score \(e(X)=P(T=1|X)\))
The main benefit is the “double protection” property: if either model is correctly specified (not necessarily both), the causal effect estimator is consistent (Funk et al. 2011).
This makes doubly robust methods especially attractive after matching or weighting, where residual imbalance may remain and model misspecification is a concern.
Why It Is “Doubly Robust”
Let:
\(T_i \in \{0,1\}\) be treatment
\(Y_i\) be outcome
\(X_i\) be pre-treatment covariates
\(\hat e(X_i)\) be estimated propensity score
\(\hat\mu_1(X_i)\) and \(\hat\mu_0(X_i)\) be estimated conditional mean outcomes
The augmented inverse probability weighted (AIPW) estimator of the ATE is:
If the outcome model is correct, residual correction terms average to zero.
If the propensity model is correct, weighting correction recovers the right estimand even if outcome regression is wrong.
A practical walkthrough with this intuition is shown in the Python Causality Handbook (Facure Alves 2022).
Link to Matching
Matching is often used as a design stage to improve overlap and comparability. Doubly robust estimation can then be used as an analysis stage on the matched sample to reduce bias from remaining imbalance.
Typical workflow:
Estimate propensity scores and match treated/control units.
Check overlap and covariate balance.
On the matched sample, fit outcome and propensity models.
Compute AIPW (or DR-ATT) estimate.
This pairing gives robust finite-sample performance in many applied settings.
Assumptions
Doubly robust methods still require core causal assumptions:
Conditional exchangeability: \[
(Y(1),Y(0)) \perp T \mid X
\]
Positivity/overlap: \[
0 < P(T=1\mid X) < 1
\]
SUTVA: no interference and well-defined treatment.
Double robustness does not solve unobserved confounding.
Estimands
ATE
Population average treatment effect (formula above).
ATT (common in matching studies)
For ATT, a doubly robust estimator can be written with treated-group focus and appropriate control reweighting/regression adjustment.
* Load dataimport delimited "lalonde.csv", clear* 1) Fit propensity modellogit treat age educ married re74 re75predict ps, prreplace ps = max(min(ps, .999999), .000001)* 2) Fit outcome models separatelyreg outcome age educ married re74 re75 if treat==1predict mu1, xbreg outcome age educ married re74 re75 if treat==0predict mu0, xb* 3) Build AIPW score and averagegen aipw_i = treat*(outcome-mu1)/ps + mu1 - (1-treat)*(outcome-mu0)/(1-ps) - mu0sum aipw_i
Diagnostics and Pitfalls
Extreme propensity scores cause unstable weights.
Use trimming or overlap checks.
Consider stabilized weights.
Poor overlap after matching can still bias estimates.
Inspect propensity distributions and common support.
Both models wrong can perform poorly.
Compare to simpler estimators (matched difference in means, regression adjustment).
Run sensitivity analyses.
Inference:
Use robust standard errors or bootstrap for confidence intervals (Funk et al. 2011).
Practical Recommendations
Use doubly robust estimation as a complement to matching, not a replacement for design diagnostics.
Keep covariates pre-treatment and theory-driven.
Report model specifications for both propensity and outcome models.
Report overlap diagnostics, balance statistics, and uncertainty method.
Summary
Doubly robust estimators are a practical way to combine matching-style design ideas with model-based correction. They provide protection against one model being misspecified and are especially useful in high-dimensional or moderately misspecified observational analyses (Funk et al. 2011; Facure Alves 2022).
Funk, Michele Jonsson, Daniel Westreich, Chris Wiesen, Til Sturmer, M. Alan Brookhart, and Marie Davidian. 2011. “Doubly Robust Estimation of Causal Effects.”American Journal of Epidemiology 173 (7): 761–67. https://doi.org/10.1093/aje/kwq439.