Imputation when only some variables a... PreviousNext
Mplus Discussion > Missing Data Modeling >
 Seth J. Schwartz posted on Monday, January 06, 2020 - 12:12 pm
Happy New Year Linda and Bengt,

I'm running a model where slopes and intercepts are imputed but the variables they are predicting are not (i.e., the outcome variables are all in one dataset). I want to combine results across the 10 imputed sets of intercepts and slopes, with the same outcome values used for each imputation. Did I read correctly in an earlier threat (from Bengt I think) that we would need to run a separate model with each set of intercepts and slopes and then manually average the point estimates and standard errors across the 10 imputations?

Thanks so much.

 Bengt O. Muthen posted on Monday, January 06, 2020 - 4:09 pm
If I understand you correctly, in the run where you do the imputations, you can add the outcome variables on the "Auxiliary" list so that they get transferred to the imputed data sets (with the same values for all sets). No manual averaging etc is needed.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message