- Statist. Sci.
- Volume 33, Number 2 (2018), 160-183.
Multiple Imputation for Multilevel Data with Continuous and Binary Variables
We present and compare multiple imputation methods for multilevel continuous and binary data where variables are systematically and sporadically missing. The methods are compared from a theoretical point of view and through an extensive simulation study motivated by a real dataset comprising multiple studies. The comparisons show that these multiple imputation methods are the most appropriate to handle missing values in a multilevel setting and why their relative performances can vary according to the missing data pattern, the multilevel structure and the type of missing variables. This study shows that valid inferences can only be obtained if the dataset includes a large number of clusters. In addition, it highlights that heteroscedastic multiple imputation methods provide more accurate inferences than homoscedastic methods, which should be reserved for data with few individuals per cluster. Finally, guidelines are given to choose the most suitable multiple imputation method according to the structure of the data.
Statist. Sci., Volume 33, Number 2 (2018), 160-183.
First available in Project Euclid: 3 May 2018
Permanent link to this document
Digital Object Identifier
Mathematical Reviews number (MathSciNet)
Audigier, Vincent; White, Ian R.; Jolani, Shahab; Debray, Thomas P. A.; Quartagno, Matteo; Carpenter, James; van Buuren, Stef; Resche-Rigon, Matthieu. Multiple Imputation for Multilevel Data with Continuous and Binary Variables. Statist. Sci. 33 (2018), no. 2, 160--183. doi:10.1214/18-STS646. https://projecteuclid.org/euclid.ss/1525313140
- Supplement to “Multiple Imputation for Multilevel Data with Continuous and Binary Variables”. Technical details on the posterior distributions of imputation model parameters and inference results for all configurations that have not been discussed in detail in the main text.
- Supplement to “Multiple Imputation for Multilevel Data with Continuous and Binary Variables”. R code for the simulation study.