Inferences on missing information under multiple imputation and two-stage multiple imputation

By Ofer Harel. 2007

86 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/18d94898e99ccce4b8caca28e3e31c57d9623f4e

How researchers see this paper

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

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

The estimation of R 2 and adjusted R 2 in incomplete data sets using multiple imputation (2009, methodology): “Due to the resemblance of the coefficient of determination to the rates of missing information (ratio of two variances) and the results in Harel [12] I imputed either 250 imputations to make sure of a stable estimate, or 25 imputations which is a commonly used number of imputations in the MI literat”
On Obtaining Estimates of the Fraction of Missing Information From Full Information Maximum Likelihood (2012, methodology): “Harel (2007) derived a standard error of the MI estimate of œj with respect to the number of imputations to better evaluate how variable the MI estimate is given a certain number of imputations.”
Pooling ANOVA Results From Multiply Imputed Datasets (2016, background): “The initial recommendation that M = 5 imputations would suﬃce for most applications of MI (Rubin, 1987) has been modiﬁed in the past by several authors (e.g., Bodner, 2008; Graham et al., 2007; Harel, 2007).”
What Improves with Increased Missing Data Imputations? (2008, methodology): “The asymptotic results of Harel (2006) demonstrate that variability of O œ m around œ is maximal for œ D :5 and approaches zero for œ near 0 and 1.”
Missing data techniques for multilevel data: implications of model misspecification (2011, background): “The efficiency of an estimate based on m imputations relative to an infinite number of imputations is (1 + γ /m)−1/2, where γ is the proportion of missing information ([24], see also [10,26]).”
How Many Imputations Do You Need? A Two-stage Calculation Using a Quadratic Rule (2016, background): “A CI for is /(cid:1839) , where z is a standard normal quantile (Harel, 2007).”