st: ice and hireg followup

[Frank Gallo <fjgallo@mac.com>]

Hi All,

I want to use -hireg- (and get its associated statistics), however, I  
have missing data: 24 variables; 3 variables (the criterion and 2  
predictors) have no missing data; complete cases 2,632 (80%; n =  
3,300); percentages missing on 2 predictors are 9.36% and 9.61%;  
percentages missing on the remaining ones range from 0.1% to 3.7%;  
total missing values 1.7% (1,338 values); the pattern is MAR. If I  
were to use -ice- and generate 5-data sets, -mim- does not allow a - 
hireg- followup. What are your thoughts on generating a single data  
set using -ice-, then run -hireg- with the robust option? Thank you.

Best,
Frank


*
*   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/