Multiple Imputation of Missing Data: A Simulation Study on a Binary Response

By J. Hardt, M. Herke, T. Brian, W. Laubach. 2013

47 citations recorded by the source corpus.

Currently, a growing number of programs become available in statistical software for multiple imputation of missing values. Among others, two algorithms are mainly implemented: Expectation Maximization (EM) and Multiple Imputation by Chained Equations (MICE). They have been shown to work well in large samples or when only small proportions of missing data are to be imputed. However, some researchers have begun to impute large proportions of missing data or to apply the method to small samples. A simulation was performed using MICE on datasets with 50, 100 or 200 cases and four or eleven variables. A varying proportion of data (3% - 63%) was set as missing completely at random and subsequently substituted using multiple imputation by chained equations. In a logistic regression model, four coefficients, i.e. non-zero and zero main effects as well as non-zero and zero interaction effects were examined. Estimations of all main and interaction effects were unbiased. There was a considerable variance in the estimates, increasing with the proportion of missing data and decreasing with sample size. The imputation of missing data by chained equations is a useful tool for imputing small to moderate proportions of missing data. The method has its limits, however. In small samples, there are considerable random errors for all effects.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/ca538e2c03a58c00b9216373466d5a633bd1b2fa

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

1. Alessandro Santaniello
2. Lorenzo Beretta

In Statistics and Probability

1. P. Madley-Dowd
2. K. Tilling
3. Alessandro Santaniello
4. Lorenzo Beretta
5. J. Heron
6. Rachael A. Hughes

Citations in the corpus

• The proportion of missing data should not be used to guide decisions on multiple imputation (2019, methodology): “Sample size has been shown to influence efficiency gains obtained via MI for binary outcomes [25] with smaller sample sizes associated with smaller gains at equivalent proportions of missing data.”
• Nearest neighbor imputation algorithms: a critical evaluation (2016, background): “The conclusions we draw applies to cases with moderate sizes of missingness, no lower than 15 % and no higher than 30 %; we intentionally limited our evaluations to this range as for small amounts of missing data, under the MAR or MCAR mechanisms, imputation may be useless and for larger amounts cau”