Donald Rubin
Harward University
Faculty of Arts and Sciences
1 Oxford Street, MA 02138
USA
Multiple imputation (MI, Rubin, 1987) was originally proposed as a method to handle missing data due to nonrepsonse in surveys, especially surveys destined to support the production of large public-use data sets. There have now been many very successful applications of MI to such data sets in the US (e.g., SCF,NHANES III, FARS, NMES), as well as to other data sets with missing data (e.g., randomized pharmaceutical trials presented to the US FDA). Recent applications also include the handling of "matrix sampling" in educational settings (e.g., NAEP) and in marketing contexts for business surveys. Even more novel applications involve the use of MI to address noncompliance in human randomized trials of anthrax vaccines and to try to build a bridge between these studies and randomized trials of macaques. In the macaque studies, true survival outcomes are measured, as well as biomarkers (e.g., blood antibody levels), whereas in the human trials, only the biomarkers are available. This presentation will be a free flowing exposition of some of these applications, and the crucial role, both conceptually and computationally, that MI makes to valid statistical analysis.
Invited presentations