Two-group logistic regression with mi... PreviousNext
Mplus Discussion > Categorical Data Modeling >
Message/Author
 Deborah Kloska posted on Friday, June 15, 2007 - 6:25 am
I want to run a two-group logistic regression model. Outcome variable has 4 levels. I discovered that I need to run a mixture model with KNOWNCLASS to get two-group results. However, when I do this (using TYPE=MIXTURE MISSING; and ESTIMATOR=MLR), I'm only obtaining complete case results.

I want to include the cases for which there are missing data on the predictors. From a previous discussion post, I tried the "trick" of introducing the predictors by specifying the variance of the predictor in the model command. I tried to put the variance statement in the %overall% section, for example

%overall%
scale on xvar;
xvar;

but then I get error messages:
*** ERROR
The following MODEL statements are ignored:
* Statements in the OVERALL class:
SCALE ON XVAR
*** ERROR
One or more MODEL statements were ignored. These statements may be
incorrect or are only supported by ALGORITHM=INTEGRATION.
(in previous runs, I've had error messages that say INTEGRATION is not an option)

I've also tried to put the variance statement for xvar on the individual class model specifications, but no luck. I've run separate logistic models for males and females, including the cases with missing data on the predictors, but my goal is to run the model as a two-group model.

Your help is appreciated.
 Linda K. Muthen posted on Friday, June 15, 2007 - 8:01 am
Please send your input, data, output, and license number to support@statmodel.com.
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: