REVIEW 6 cited by
An Introduction to Proximal Causal Learning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
A standard assumption for causal inference from observational data is that one has measured a sufficiently rich set of covariates to ensure that within covariate strata, subjects are exchangeable across observed treatment values. Skepticism about the exchangeability assumption in observational studies is often warranted because it hinges on investigators' ability to accurately measure covariates capturing all potential sources of confounding. Realistically, confounding mechanisms can rarely if ever, be learned with certainty from measured covariates. One can therefore only ever hope that covariate measurements are at best proxies of true underlying confounding mechanisms operating in an observational study, thus invalidating causal claims made on basis of standard exchangeability conditions. Causal learning from proxies is a challenging inverse problem which has to date remained unresolved. In this paper, we introduce a formal potential outcome framework for proximal causal learning, which while explicitly acknowledging covariate measurements as imperfect proxies of confounding mechanisms, offers an opportunity to learn about causal effects in settings where exchangeability on the basis of measured covariates fails. Sufficient conditions for nonparametric identification are given, leading to the proximal g-formula and corresponding proximal g-computation algorithm for estimation. These may be viewed as generalizations of Robins' foundational g-formula and g-computation algorithm, which account explicitly for bias due to unmeasured confounding. Both point treatment and time-varying treatment settings are considered, and an application of proximal g-computation of causal effects is given for illustration.
Forward citations
Cited by 6 Pith papers
-
The Spectral Structure of Latent Treatment Effects
After shared-subspace compression, the difference of treatment-arm proxy quotient operators is similar to the diagonal of latent treatment effects, whose eigenvalues and lifted eigenvectors recover the full mixture.
-
Proximal Mediation Analysis with Unmeasured Treatment-Induced Confounding
Interventional mediation effects are identifiable under unmeasured treatment-induced confounding given two valid proxies, via four proximal formulas and a multiply robust, locally efficient estimator.
-
Quantile-Optimal Policy Learning under Unmeasured Confounding
Under instrumental-variable or negative-control assumptions, the authors prove a pessimism-based policy learning method achieves about 1/sqrt(n)-type regret for quantile reward objectives with unmeasured confounders.
-
Conditioning-Depth Diagnostics for Hidden Memory in Temporal Causal Discovery
Depth-dependent instability of inferred causal graphs is proposed as a non-specific warning sign of hidden memory in time-series causal discovery, demonstrated on simulations and four zebrafish calcium-imaging recordings.
-
Source-Condition Analysis of Kernel Adversarial Estimators
A statistics abstract about finite-sample bounds for kernel adversarial estimators is joined in the submission to the complete text of a different astronomy paper, so the claimed results have no supporting body.
-
From Observational Data to Clinical Recommendations: A Causal Framework for Estimating Patient-level Treatment Effects and Learning Policies
A causal framework for learning treatment policies with deferral, applied to diuretic dosing in acute heart failure with kidney injury.
Discussion (0). Continue with ORCID to comment.