Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: two multiple imputation questions

From   "Ploutz-Snyder, Robert (JSC-SK)[USRA]" <>
To   "" <>
Subject   st: two multiple imputation questions
Date   Fri, 2 Apr 2010 13:33:07 -0500


I have a data set with multiple predictors (X's) and multiple outcomes (Y's=) and am interested in using the multiple imputation functions (mi) in Stat a to help fill in some missing data.

One of the analyses that I wish to do after running mi is  seemingly unrelated regression, since we have multiple correlated Y's, predicting all from the same set of X's.  The general syntax is mi estimate: blah... where blah=  is the estimation command of interest.  Unfortunately,  sureg is not one of the supported estimated commands in mi.

I see that there is an option to mi estimate, cmdok, which "allows unsupported estimation commands to be used with mi estimate," but I'm not exactly sure what it means to allow an unsupported command??  Is this a valid way to  go about sureg on a sample that used mi??  Will this option inform the model about the multiple imputations appropriately??

Thanks for any assistance..


*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index