On the bias of the multiple‐imputation variance estimator in survey sampling

By Jae Kwang Kim, J. Michael Brick, Wayne A. Fuller, Graham Kalton. 2006

127 citations recorded by the source corpus.

Subfields: Sociology and Political Science, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/6dd88730e83c6b42fc83b454f0e7ed9cd490db9c

How researchers see this paper

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

In Sociology and Political Science

In Statistics and Probability

Citations in the corpus

Combining Multiple Imputation and Inverse-Probability Weighting (2012, background): “Proofs of parts (i) and (ii) come from extending the proof of Kim et al. (2006), which shows (ii) is true in the special case where Z = 1; that of part (iii) comes from applying Theorem 2 of Robins and Wang (2000).”
A unified approach to linearization variance estimation from survey data after imputation for item nonresponse (2009, background): “Fay (1992) and Kim et al. (2006) discussed situations where the congeniality assumption does not hold and thus WM + BM can be biased for var(θ̂I d ).”
Review of inverse probability weighting for dealing with missing data (2013, methodology): “In such situations, it may be worth combining IPW and MI, imputing missing values in individuals with almost complete data and using IPW to adjust for the exclusion of individuals with more missing data.(7,16,18,19,33,34) This approach offers some of the efficiency advantages of MI while minimising ”
Calibrated Bayes, an alternative inferential paradigm for official statistics (2012, background): “Multiple imputation does not yield consistent estimates of variance under particular forms of model misspecification (Meng 1994; Rao 1996; Fay 1996; Robins and Wang 2000; Kim et al., 2006).”
lavaan.survey: An R Package for Complex Survey Analysis of Structural Equation Models (2014, background): “It should be noted, however, that any survey weights should be included in the imputation models, and Equation (3) may not consistently estimate the variance if the response mechanism is not at random with respect to the weights (Kott 1995; Kim, Brick, Fuller, and Kalton 2006).”
Flexible Imputation of Missing Data (2012, background): “Several authors have shown that Rubin’s estimate of the variance is biased (Wang and Robins, 1998; Robins and Wang, 2000; Nielsen, 2003; Kim et al., 2006).”