Search found 49 matches

by Harvey Goldstein
Thu May 28, 2015 3:17 pm
Forum: Realcom user forum
Topic: Realcom-imput / imputation model with missing values in explanatory variables
Replies: 3
Views: 5379

Re: Realcom-imput / imputation model with missing values in explanatory variables

You should read the manual that is on the website. It does not used chained equations and you can spcify missing data anywhere and imputed values are available.
Harvey Goldstein
by Harvey Goldstein
Sat Feb 14, 2015 1:15 pm
Forum: Realcom user forum
Topic: Imputation and perfect prediction
Replies: 2
Views: 3485

Re: Imputation and perfect prediction

We haven't really explored this. There are some bounds that are placed around draws that may deal with this, but you would do best to avoid perfect predictions if you can. Interested to know if it works for you however.
Harvey Goldstein
by Harvey Goldstein
Sat Feb 14, 2015 2:03 am
Forum: Realcom user forum
Topic: Include level1 variables to impute level 2 missing variable
Replies: 1
Views: 3356

Re: Include level1 variables to impute level 2 missing variable

When you specify a 2 level imputation model any level 1 variables that are declared responses at level 1 implicitly contribute through their level 2 random effects. However you cannot have a level 1 predictor for a level 2 response in any multilevel model so do not include level 1 variables as covar...
by Harvey Goldstein
Fri Dec 12, 2014 6:18 pm
Forum: Realcom user forum
Topic: Imputation and perfect prediction
Replies: 2
Views: 3485

Re: Imputation and perfect prediction

There is no feature in realcom to do this I'm afraid.
Harvey Goldstein
by Harvey Goldstein
Wed Sep 03, 2014 10:32 am
Forum: MLwiN user forum
Topic: Missing data: FIML and (multilevel) multiple imputation
Replies: 1
Views: 3509

Re: Missing data: FIML and (multilevel) multiple imputation

The IGLS estimates are indeed maximum likelihood ones. Re missing data the mlwin default is listwise deletion of all level 1 records where any model variable has a missing value. If you wish to do an efficient multiple imputation on 2-level data you have two possibilities now. Either REALCOM which h...
by Harvey Goldstein
Mon Jun 16, 2014 1:34 pm
Forum: Realcom user forum
Topic: model with level 2 missings only
Replies: 3
Views: 5032

Re: model with level 2 missings only

Should be original 2 level dataset
Harvey
by Harvey Goldstein
Sat Jun 14, 2014 8:44 pm
Forum: Realcom user forum
Topic: model with level 2 missings only
Replies: 3
Views: 5032

Re: model with level 2 missings only

Yes - that should be fine. Let me know if there are any problems.
Harvey Goldstein
by Harvey Goldstein
Thu Apr 10, 2014 2:17 pm
Forum: Realcom user forum
Topic: Imputation very slow
Replies: 7
Views: 7911

Re: Imputation very slow

Yes - imputation can be slow. We are switching now to STATJR (see CMM web site) and a 2-level imputation module has just been put up there - use of this is free.
You should ideally put all variables in as responses.
Harvey Goldstein
by Harvey Goldstein
Mon Mar 03, 2014 2:52 pm
Forum: Realcom user forum
Topic: 2 level vs. single level imputation model
Replies: 6
Views: 6780

Re: 2 level vs. single level imputation model

All variables in the MOI, whether responses or predictors do need to be in the imputation model.
by Harvey Goldstein
Mon Mar 03, 2014 2:00 pm
Forum: Realcom user forum
Topic: 2 level vs. single level imputation model
Replies: 6
Views: 6780

Re: 2 level vs. single level imputation model

Strictly speaking you should set up a 2-level imputation model where any level 1 vbles in the MOI are responses (at level 1 in the imputation model).
Harvey Goldstein