Dear all,
I'm using ice to create an imputed dataset. The outcome I'm interested in is a 30% or more change in 3 variables over 2 time points. And I have a series of other variables that are the predictors (e.g. v1 v2 v3 v4 v5 v6)
However, I am unsure what I should specify my ice equation. Would it be best to include the 30% change(x30) in the ice equation. For example: ice x30 v1 v2 v3 v4 v5 v6......
Or should I include the raw variables from which the 30% change is calculated (x1 x2), and then after creating the imputed dataset calculate the 30% change
Eg. ice x1 x2 v1 v2 v3 v4 v5 v6.....
Any thoughts appreciated
Sham
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
