st: MI estimate with meologit using Stata 13 SE x64

[Mark McCann <mark.mccann@qub.ac.uk>]

Hello,

I've just installed Stata 13 and I'm trying to account for missing data in a multilevel ordered logistic regression. I'm not even sure if this is feasible but here's what I've tried.


.mi estimate: meologit s5alc s4alc y4mon ||school: , or

.mi estimate, cmdok : meologit s5alc s4alc y4mon ||school: , or

	these both return the error

command name required
r(100);

mi estimate, cmdok : gsem (s5alc <- s4alc y4mon M1[school]) , ologit

gives the error 

name conflict
                 stata():  3598  Stata returned error
             _e_repost():     -  function returned error
                 <istmt>:     -  function returned error
r(3598);


I also see that the gsm doesn't allow for full information estimation.

The MI imputation was performed using the realcomImpute and realcomImputeLoad commands.

Has anybody any advice? Is this something that won't be supported until future versions of Stata, or are there any other potential fixes in the meantime?


One other minor thing. The do file editor appears to have a vertical line 80 characters across. Is there any explanation for this?

Kind Regards,
		Mark


Research Statistician
Institute of Child Care Research
Queen's University Belfast
6 College Park
Belfast BT7 1LP

Tel: +44(0)28 9097 3163

Facebook.com/BYDSNI 
blogs.qub.ac.uk/griefstudy/
Twitter: @Mark_ICCR 



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/