3 step approach of GMM PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 TaeKyoung Lee  posted on Saturday, May 02, 2015 - 8:46 am
Dear Drs. Muthens,

I have two questions regarding auxiliary option in GMM.

First, using AUXILIARY option(three step approach), I am now analyzing conditional GMM. Regarding auxiliary variables, my first question is whether Mplus allows to specify auxiliary variables as a latent variable.

Second, when applying to 3-step approach (R3STEP) for predictors and/or Lanza's two methods ((DCAT) and (DCON) options) for distal outcomes, I detected missing cases were automatically deleted from the covariate analyses (Multinomial regression for predictors and Mean Equality test for distal outcomes, respectively). However, when applying to another 3-step approach for distal outcomes such as (DU3STEP) or (DE3STEP), I couldn't find any missing cases issues in the results of Mean equality test. So, if I run the GMM in a correct way, my second question is why these missing cases differently influence the 3-STEP approach. Can I ask you for your advice regarding?

I am more than grateful for your suggestion.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: