Multiple imputation of missing values: Further update of ice, with an emphasis on categorical variables

Abstract

Multiple imputation of missing data continues to be a topic of considerable interest and importance to applied researchers. In this article, the ice package for multiple imputation by chained equations (also known as fully conditional specification) is further updated. Special attention is paid to categorical variables. The relationship between ice and the new multiple-imputation system in Stata 11 is clarified. © 2009 StataCorp LP.