Page 1 of 1

Level 2 imputation: MISSINGFULL error

Posted: Wed Aug 23, 2017 11:18 am
by gilleecl

Thanks to your earlier help, I have got an imputation model running when imputing Level 1 variables only. However, can you advise on the imputation of level 2? When trying a test MI model, with one L2 response variable and two auxiliary variables (cons and another L2 var with no missing vals), I get an error message about accessing MISSINGFULL (see attached). Would you have any advice? I have organised the input file so that all Level 1 vars are before Level 2 variables and the file is sorted by level 2 and level 1 IDs.
Many thanks

Re: Level 2 imputation: MISSINGFULL error

Posted: Tue Sep 12, 2017 4:00 pm
by ChrisCharlton
Would it be possible to provide some more information regarding your model specification? In particular could you provide a screenshot of the equation displayed by Realcom as this should provide us with additional details? If you were able to provide any example data with the steps that cause the error then we can also attempt to replicate it.

Re: Level 2 imputation: MISSINGFULL error

Posted: Mon Nov 20, 2017 6:30 am
by RitaPickett
MISSING FULL error comes more frequently with imputation models. As chris told it would be easier if you had provided with a screen shot. If so we could encounter error and replace it earlier. Use Multiple Imputation. Multiple Imputation fills in estimates for the missing data. But to capture the uncertainty in those estimates, MI estimates the values multiple times. Because it uses an imputation method with error built in, the multiple estimates should be similar, but not identical. Otherwise you can analyze the full, incomplete data set using maximum likelihood estimation. This method does not impute any data, but rather uses each cases available data to compute maximum likelihood estimates.

Re: Level 2 imputation: MISSINGFULL error

Posted: Tue Nov 21, 2017 9:10 am
by gilleecl
thanks, I had found a work around at the time but once I get back to this project, i'll certainly provide a screenshot if I can replicate the error. thanks for your help.