|
|
LCGA R3STEP Imputation of auxiliaries |
|
Message/Author |
|
Jan Hoeltge posted on Thursday, July 04, 2019 - 6:04 pm
|
|
|
Hello, some questions: 1) Is it ok to use R3STEP to predict class membership in LCGA? 2) I ran the analysis and it worked, but I have some missings and the analysis uses listwise deletion. However, it recommends to the following: DATA IMPUTATION CAN BE USED FOR THE AUXILIARY VARIABLES FOLLOWED BY ANALYSIS WITH TYPE=IMPUTATION But I use ANALYSIS: TYPE = MIXTURE to do the analysis and TYPE = IMPUTATION MIXTURE does not work. So where do I need to put the impute command and should it rather look like (from Chapter 11: Missing Data Modeling and Bayesian Analysis): DATA IMPUTATION: IMPUTE = y1-y4 x1 (c) x2; NDATASETS = 10; Thanks, Jan |
|
|
(1) Yes. (2) The idea is to first do imputation and then read in Type=Imputation via the Data command for the mixture analysis. So 2 steps. |
|
Back to top |
|
|