Correcting for non-compliance in randomized trials using structural nested mean models

By James M. Robins. 1994

Subfields: Statistics and Probability, Causal Inference & Discovery

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

In Causal Inference & Discovery

Citations in the corpus

Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounders (2019, methodology): “Standard methods for performing counterfactual inference in longitudinal data are found in the epidemiology literature and include the g-computation formula, gestimation of structural nested mean models, and inverse probability of treatment weighting of marginal structural models (Robins, 1994; Robi”
G-Net: A Deep Learning Approach to G-computation for Counterfactual Outcome Prediction Under Dynamic Treatment Regimes (2020, methodology): “G-methods include gcomputation [4–6], structural nested models [7, 8], and marginal structural models [9, 10].”
Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representations (2020, methodology): “Standard methods for estimating the effects of time-varying exposures were ﬁrst developed in the epidemiology literature and include the g-computation formula, Structural Nested Models and Marginal Structural Models (MSMs) (Robins, 1986; 1994; Robins et al., 2000; Robins & Hernán, 2008).”
Learning When-to-Treat Policies (2019, background): “, 1992) learns the Structural Nested Mean Models (SNMM) (Robins, 1994) which model the difference in the marginal outcome functions directly.”
Using Decision Lists to Construct Interpretable and Parsimonious Treatment Regimes (2015, methodology): “Examples include estimators of the regression of an outcome on covariates, treatment, and their interactions (e.g., Su et al., 2009; Qian and Murphy, 2011; Tian et al., 2014), and estimators of point treatment effect given covariates (e.g., Robins, 1994; Vansteelandt et al., 2014).”
Marginal and Nested Structural Models Using Instrumental Variables (2010, methodology): “The purpose of this article is to extend the structural mean models of Robins (1989, 1994) and Vansteelandt and Geotghebeur (2003) and develop corresponding estimation methods.”
Estimation and extrapolation of optimal treatment and testing strategies (2008, background): “To overcome this difficulty, following Robins [9] we can consider the semilinear general MSM E[Ydx |V ]=h1(x,V, ∗1)+h∗2(V ) with h∗2(V ) allowed to be an arbitrary unknown function, so as to prevent bias in the estimation of h1(x,V, ∗ 1) from misspecification of a parametric model for h ∗ 2(V ).”
A confounding bridge approach for double negative control inference on causal effects (2018, unknown): “The instrumental variable (IV) approach (Angrist et al., 1996; Baker & Lindeman, 1994; Goldberger,1972; Robins, 1994; Wright, 1928), rests on an auxiliary covariate that (i) has no direct effect on the outcome, (ii) is independent of the unmeasured confounder, and (iii) is associated with the…”
Structural nested models and G-estimation: the partially realized promise (2014, methodology): “Causal effects can now be defined as comparisons of potential outcomes Y a and Y a † for the same individual subject or group of subjects for different treatments a and a† (Rubin (1978); Robins (1986)).”
DOUBLY ROBUST INSTRUMENTAL VARIABLE REGRESSION (2012, methodology): “Robins (1994) shows how one can estimate parameters of interest by modeling E(Zi|Xi) in structural nested mean models which include our model as a special case.”
Instruments for Causal Inference: An Epidemiologist's Dream? (2006, background): “EY(1) EY(0) EYX 0 {1 EX} exp (0 * ) 1 EX EYX 1 ( 10 )”
Instrumental Variables: An Econometrician's Perspective (2014, unknown): “Examples of this type are considered in Zhang, Rubin, and Mealli (2009), Frumento, Mealli, Pacini, and Rubin (2011), and Robins (1986). Suppose”
Constructing Dynamic Treatment Regimes in Infinite-Horizon Settings (2014, unknown): “: S t+1(A t) = S t+1 for each t 2. Sequential randomization: For each t, A tis conditionally independent of S t+1(a t) given S t. These assumptions link the potential outcome and the observed data (Robins, 1994, 1997). Assumption 1 means that the potential outcome of a treatment regime corresponds t”
Optimal a priori balance in the design of controlled experiments (2013, background): “Freedman (1997, 2008), Robins (1994, 2004) and Berk (2004) have drawn attention to the fact that, when such a parametric modelling assumption is false, this estimate may be extremely inaccurate and may lead to false conclusions.”
Bias-Reduced Doubly Robust Estimation (2015, methodology): “Robins et al. (1994) showed that consistent estimation of this mean outcome μ0 requires specification of at least one of the two following working models.”
Tutorial in Biostatistics: Instrumental Variable Methods for Causal Inference* (2014, background): “A weaker restriction on the heterogeneity of treatment effects among the compliance classes is the no current treatment value interaction assumption [47]:”