Estimation of Regression Coefficients When Some Regressors are not Always Observed

By James M. Robins, Andrea Rotnitzky, Lue Ping Zhao. 1994

How researchers see this paper

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

In Statistics and Probability

Citations in the corpus

Balanced Policy Evaluation and Learning (2017, methodology): “[13] propose a weighting-based approach that combines the direct and IPW approaches by adapting the doubly robust (DR) estimator [11, 34, 35, 38]: ⌧̂DR(⇡) = 1 n Pn i=1 Pm t=1 ⇡t(Xi)μ̂t(Xi) + 1 n Pn i=1(Yi μ̂Ti(Xi))⇡Ti(Xi)/'̂Ti(Xi).”
Doubly Robust Policy Evaluation and Learning (2011, methodology): “Furthermore, most of previous analysis of doubly robust estimation studies asymptotic behavior or relies on various modeling assumptions (e.g., Robins et al. (1994), Lunceford & Davidian (2004), and Kang & Schafer (2007)).”
Efficiently Breaking the Curse of Horizon in Off-Policy Evaluation with Double Reinforcement Learning (2019, background): “When γ = 0, this reduces to the bound in the no-horizon bandit OPE setting (Robins et al. 1994):”
Intrinsically Efficient, Stable, and Bounded Off-Policy Evaluation for Reinforcement Learning (2019, methodology): “The third approach is the doubly robust (DR) method, which combines DM and IS and is given by adding the estimated Q-function as a control variate [5, 7, 25].”
Learning When-to-Treat Policies (2019, background): “This estimator generalizes the well known augmented inverse propensity weighted estimator of Robins, Rotnitzky, and Zhao (1994) beyond the static case.”
More Robust Doubly Robust Off-policy Evaluation (2018, methodology): “On the other hand, the semi-parametric theory in multivariate regression (Robins et al., 1994) states that without extra assumption on the data distribution, the class of unbiased, consistent, and asymptotically normal OPE estimators is asymptotically equivalent to the DR estimators in (7).”
Comment: Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data. (2008, methodology): “Under ignorability, the πi’s play no role in likelihood-based or Bayesian analyses for the parameters of the y-model (Rubin, 1976). The parametric approach—specifying a full model for yi and ignoring the πi’s—is emphasized in texts on missing data by Rubin (1987), Schafer (1997), Little and Rubin (2”
Representation Balancing MDPs for Off-Policy Policy Evaluation (2018, background): “[17] J.”
A confounding bridge approach for double negative control inference on causal effects (2018, unknown): “hold in certain causal inference problems (Robins et al., 1994; Van der Laan & Robins, 2003; Bang & Robins, 2005; Tchetgen Tchetgen et al., 2010).”
Strati cation and weighting via the propensity score in estimation of causal treatment e ects : a comparative study (2004, background): “However, they are special cases of a broader class of estimators that may be deduced by viewing the situation as a ‘missing data’ problem discussed in a landmark paper by Robins, Rotnitzky, and Zhao [13].”
CAUSAL INFERENCE WITH A GRAPHICAL HIERARCHY OF INTERVENTIONS (2014, methodology): “This estimator is combined with a similarly defined doubly robust estimator for (a, a) derived in [20] to yield a triply robust estimator for the direct and indirect parameters on the difference scale.”
The performance of estimators based on the propensity score (2013, methodology): “The third class of direct matching estimators considered is the one-to-many radius (or caliper) matching algorithm as, for example, discussed by Rosenbaum and Rubin (1985) and used by Dehejia and Wahba (1999, 2002).”
Estimating Equations Inference With Missing Data (2008, methodology): “Robins et al. (1994) suggested that one might estimate P(x) by modeling the dependence of δ on (Yi,Zi ) via a parametric model of some sort, such as logistic regression.”