Although multiple imputation is the gold standard of treating missing data, single ratio imputation is often used in practice. Based on Monte Carlo simulation, the Expectation-Maximization with Bootstrapping (EMB) algorithm to create multiple ratio imputation is used to fill in the gap between theory and practice.




While this article appears in 16(1) in the section Algorithms and Code, it was submitted and accepted to this issue's Regular Articles, and the placement is an oversight of the Editors. For reasons of pagination, reassignment of the article is problematic; however, it received a full double-blind peer review before acceptance, as do all Regular Articles in JMASM.

Recommended Citation

Takahashi, M. (2017). Multiple ratio imputation by the EMB algorithm: theory and simulation. Journal of Modern Applied Statistical Methods, 16(1), 630-656. doi: 10.22237/jmasm/1493598840