What is the difference between these two procedures? As help
description indicates ICE uses switching regression to impute values.
How does this differ from what MI impute does (say for regression for
a continuous variable)? I see there is an option to check the
stability of MCMC (I assume this is referring to the stationarity of
the posterior distribution estimated using MCMC?). From what I can
tell, ICE does not use MCMC, is this the main difference between the
two?
Thanks for your help,
Woolton
*
* 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/
