A comparison of multiple imputation and doubly robust estimation for analyses with missing data

By James R. Carpenter, Michael G. Kenward, Stijn Vansteelandt. 2006

256 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/1466fee22d33bbfdd4117e83e15f81708e4331c3

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

Estimating flight-level price elasticities using online airline data: A first step toward integrating pricing, demand, and revenue optimization (2014, methodology): “However, this method will typically lead to invalid and biased results (Carpenter et al., 2006).”
Weighted Generalized Estimating Functions for Longitudinal Response and Covariate Data That Are Missing at Random (2010, unknown): “Further evaluations of DR estimators have been reported by Lunceford and Davidian (2004), Carpenter, Kenward, and Vansteelandt (2006), Davidian, Tsiatis, and Leon (2005), Bang and Robins (2005), and Kang and Schafer (2007).”
Efficient and Doubly Robust Imputation for Covariate-Dependent Missing Responses (2008, methodology): “Carpenter, Kenward, and Vansteelandt (2006) compared some traditional imputation methods with the Robins–Rotnitzky–Zhao inverse probability weighting method and found that the latter is preferable, because it is more efficient than a nonparametric imputation method and more robust than a parametric…”
Causal mediation analysis for stochastic interventions (2019, methodology): “tandard data analytic practice, we advocate for the use of the proposed methodology in tandem with a correction for missing data, such as imputation or weighting by inverse probability of censoring. (Carpenter et al., 2006;Vansteelandt et al.,2010;Seaman et al.,2012). To demonstrate the assessment o”
Comparison of Four Methods for Handing Missing Data in Longitudinal Data Analysis through a Simulation Study (2014, background): “MNAR is particularly useful in assessing the sensitivity of the results that are not MAR [5] and it is highly recommended to be incorporated into the analysis.”
Bayesian Nonparametric Modeling for Causal Inference (2011, background): “Estimators with this property are called “doubly robust” or “doubly protected” and are discussed further by Carpenter, Kenward, and Vansteelandt (2005) and Kang and Schafer (2007).”
A Semiparametric Estimation of Mean Functionals With Nonignorable Missing Data (2011, methodology): “The pseudo-values in (17) bear the same form as those in Carpenter, Kenward, and Vansteelandt (2006) under ignorable missing, and are also used for variance estimation in Kim and Rao (2009).”
Much Ado About Nothing (2007, methodology): “Another approach to accounting for missing predictor data is the use of weighting methods (Robins et al. 1995; Xie and Paik 1997; Horton and Lipsitz 1999; Horton et al. 2001; Carpenter et al. 2006).”