Little research has been devoted to multiple imputation (MI) of derived variables. We investigated various MI approaches for the outcome, rate of change, when the analysis model is a two-stage linear regression. Our simulations showed that competitive approaches depended on the missing data mechanism and presence of auxiliary terms.
Desai, Manisha; Mitani, Aya A.; Bryson, Susan W.; and Robinson, Thomas
"Multiple Imputation When Rate of Change Is The Outcome of Interest,"
Journal of Modern Applied Statistical Methods: Vol. 15
, Article 10.
Available at: http://digitalcommons.wayne.edu/jmasm/vol15/iss1/10