Semiparametric Proximal Causal Inference
نویسندگان
چکیده
Skepticism about the assumption of no unmeasured confounding, also known as exchangeability, is often warranted in making causal inferences from observational data; because exchangeability hinges on an investigator’s ability to accurately measure covariates that capture all potential sources confounding. In practice, most one can hope for covariate measurements are at best proxies true underlying confounding mechanism operating a given study. this paper, we consider framework proximal inference introduced by Miao et al. (2018); Tchetgen (2020), which while explicitly acknowledging imperfect mechanisms, offers opportunity learn effects settings where basis measured fails. We make number contributions including (i) alternative set conditions nonparametric identification average treatment effect; (ii) general semiparametric theory estimation effect efficiency bounds key models interest; (iii) characterization doubly robust and locally efficient estimators effect. Moreover, provide analogous results treated. Our approach illustrated via simulation studies data application evaluating effectiveness right heart catheterization intensive care unit critically ill patients.
منابع مشابه
Cp criterion for semiparametric approach in causal inference
For marginal structural models, which recently play an important role in causal inference, we consider a model selection problem in the framework of a semiparametric approach using inverse-probability-weighted estimation or doubly robust estimation. In this framework, the modeling target is a potential outcome which may be a missing value, and so we cannot apply the AIC nor its extended version...
متن کاملSemiparametric Inference and Models
We review semiparametric models and various methods of inference – efficient and inefficient – for such models.”
متن کاملCausal Inference
This paper reviews a theory of causal inference based on the Structural Causal Model (SCM) described in (Pearl, 2000a). The theory unifies the graphical, potential-outcome (NeymanRubin), decision analytical, and structural equation approaches to causation, and provides both a mathematical foundation and a friendly calculus for the analysis of causes and counterfactuals. In particular, the paper...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Journal of the American Statistical Association
سال: 2023
ISSN: ['0162-1459', '1537-274X', '2326-6228', '1522-5445']
DOI: https://doi.org/10.1080/01621459.2023.2191817