Assessing Sensitivity to an Unobserved Binary Covariate in an Observational Study with Binary Outcome

By Paul R. Rosenbaum, Donald B. Rubin. 1983

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Counterfactual Prediction with Deep Instrumental Variables Networks (2016, background): “ed Instrument: z is conditionally independent of the error – i.e., z ??e jx.3 The last condition is chosen to match the ‘unconfoundedness’ assumption in the Neyman-Rubin potential outcomes framework [Rosenbaum and Rubin, 1983] (i.e. p ??e jx). But our assumption is weaker — in particular, we allow f”
Simple Rules for Complex Decisions (2017, methodology): “We address it here by adapting tools from causal inference to the policy seing, including the method of Rosenbaum and Rubin [29] for assessing the sensitivity of estimated causal eects to an unobserved binary covariate.”
Simple rules to guide expert classifications (2020, methodology): “We investigated three approaches to estimating V π—response surface modelling (Hill, 2012), inverse propensity weighting (Rosenbaum and Rubin, 1983b, 1984) and doubly robust estimation (Cassel et al., 1976; Robins et al., 1994; Robins and Rotnitzky, 1995; Kang and Schafer, 2007; Dudı́k et al.,…”
Minimizing Sensitivity to Model Misspecification (2018, background): “…2016, Hansen and Marinacci, 2016, and especially Hansen and Sargent, 2008) and the literature on sensitivity analysis in statistics and economics (e.g., Rosenbaum and Rubin, 1983, Leamer, 1985, Imbens, 2003, Altonji et al., 2005, Nevo and Rosen, 2012, Oster, 2019, Masten and Poirier, 2020, 2021).”
Teaching Statistical Inference for Causal Effects in Experiments and Observational Studies (2004, background): “This is also known as the assumption of “strongly ignorable treatment assignment” (Rosenbaum & Rubin, 1983a).”
The sensitivity of linear regression coefficients' confidence limits to the omission of a confounder (2009, methodology): “Sensitivity analysis methods for various models and data structures are proposed in Cornfield et al. (1959), Rosenbaum and Rubin (1983), Rosenbaum (1988), Copas and Li (1997), Robins, Rotnitzky and Scharfstein (2000), Scharfstein and Irizarry (2003), Marcus (1997), Lin, Psaty and Kronmal (1998),…”
Dealing with limited overlap in estimation of average treatment effects (2009, unknown): “To solve the identification problem, we maintain throughout the paper the unconfoundedness assumption (Rubin, 1978; Rosenbaum and Rubin, 1983a), which asserts that conditional on the pre-treatment variables, the treatment indicator is independent of the potential outcomes: Assumption 3.1…”
Principal stratification analysis using principal scores (2016, methodology): “l studies without access to the outcome data, which prevents data snooping by selecting favorable outcome models. In parallel with this classical wisdom of propensity scores in observational studies (Rosenbaum and Rubin 1983b), we propose to conduct principal stratication analysis based on principal”
Using Propensity Scores to Help Design Observational Studies: Application to the Tobacco Litigation (2001, methodology): “The central result in Rosenbaum and Rubin (1983a) is that if a group of treated units and control units have the same value of the propensity score, ei, then they have the same distribution of multivariate Xi, no matter what the dimension of Xi.”
Design sensitivity in observational studies (2004, methodology): “Alternative methods of sensitivity analysis for hidden bias in observational studies are discussed by Cornfield et al. (1959), Rosenbaum & Rubin (1983), Gastwirth (1992), Copas & Li (1997), Lin et al. (1998), Robins et al. (1999), Copas & Eguchi (2001) and Imbens (2003).”
Bias From Classical and Other Forms of Measurement Error (2001, methodology): “values consistent with the amount of measurement error under the various models, as we illustrate in Section 4. These analyses are in the spirit of the sensitivity and bounds analyses of Rosenbaum and Rubin (1983) , Leamer (1987), Horowitz and Manski (1995), and Bollinger (1996).”