Multiple Imputation and its Application

By James R. Carpenter, Michael G. Kenward. 2013

642 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/9e0c151119463e051a7863ecf0ed8df373061e22

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

LinDA: linear models for differential abundance analysis of microbiome compositional data (2021, methodology): “Once we obtain the posterior samples, LinDA can be applied to the posterior samples and results are then aggregated, similar in the spirit to the multiple imputation method [36].”
Missing data and multiple imputation in clinical epidemiological research (2017, unknown): “Auxiliary variables that are strongly associated with both the value and the missingness are more likely to have an impact on the results of multiple imputation and reduce bias.(19) Based on our knowledge of the data, research question, or literature, we may Figure 4 Selection of variables in order ”
Accounting for missing data in statistical analyses: multiple imputation is not always the answer (2019, methodology): “However, exploratory analyses can refute MCAR by identifying observed predictors of the missingness mechanism.(19) For example, distributions of variables for socioeconomic status can be compared between participants with observed and missing tobacco use.”
Multiple imputation as a flexible tool for missing data handling in clinical research. (2017, methodology): “This approach is sometimes referred to as latent variable imputation (Carpenter & Kenward, 2013; Enders, Mistler, & Keller, 2016; Quartagno & Carpenter, 2016) because it views discrete responses as arising from an underlying normal latent variable distribution (or distributions of latent variable…”
Pooling ANOVA Results From Multiply Imputed Datasets (2016, methodology): “Table 1 provides an because the exact missing data mechanism need not be known in order to perform MI. Treating nonignorable missing data requires making strong assumptions about the missing data mechanism and thus was not considered in this study (see Carpenter & Kenward, 2013).”
Attrition in developmental psychology (2017, methodology): “Under plausible MNAR conditions, however, the consistency of inferences from results can be assessed to provide insight on their trustworthiness (see Carpenter & Kenward, 2013; Enders, 2010, sections 10.16–10.20).”
Thresholds for statistical and clinical significance in systematic reviews with meta-analytic methods (2014, methodology): “For all meta-analyses, we recommend using at least two sensitivity analyses to assess the potential impact of the missing outcome data (risk of attrition bias) on the meta-analysis results [67].”
Multiple Imputation of Missing Data for Multilevel Models (2018, methodology): “…approach is different, it has been argued that the two approaches imply similar covariance structures at the individual and the group level and can be used interchangeably (e.g., Carpenter & Kenward, 2013, p. 220; Lüdtke et al., 2017; Mistler, 2015; however, see also Resche-Rigon & White, 2016).”
Subject specific and population average models for binary longitudinal data: a tutorial (2013, methodology): “More generally, weighted GEE estimation can be performed to allow MAR data, and multiple imputation methods can be used for either approach to ‘fill in’ incomplete data sets under the MAR assumption (Carpenter & Kenward, 2013).”
Fitting multilevel multivariate models with missing data in responses and covariates that may include interactions and non‐linear terms (2014, background): “Furthermore, in practice, imputed data from these two alternatives are practically indistinguishable unless the fitted probabilities are very close to 0 or 1 (Carpenter and Kenward (2013), page 95).”
missMDA: A Package for Handling Missing Values in Multivariate Data Analysis (2016, methodology): “For more details on these methods, we refer the reader to Little and Rubin (1987, 2002), Van Buuren (2012) and Carpenter and Kenward (2013).”