Auxilary Vars. to make data MAR PreviousNext
Mplus Discussion > Missing Data Modeling >
Message/Author
 Nina L. Alesci posted on Monday, April 21, 2008 - 2:49 am
Graham (2003) and Enders(2006) suggest code to make data MAR:

USEVARIABLES are stage1-stage10 eversmk1 livwsmkr racewhte;
MISSING = all (-1234);
Analysis:
TYPE = MISSING;
Model:
i s| stage1@0 stage2@.5 stage3@1 stage4@1.5 stage5@2 stage6@2.5 stage8@3.5 stage9@4 stage10@4.5;
i on eversmk1;
[stage1-stage10@0];
[i s];
stage1-stage10;
i s;
!auxiliary variables;
livwsmkr racewhte;
!correlations among AVs;
livwsmkr with racewhte;
!correlate with IVs;
livwsmkr racewhte with eversmk1;
!correlate with DVs;
livwsmkr racewhte with stage1-stage10;

Output:
sampstat standardized;

This message appears for binary or continuous aux vars.
WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) IS NOT POSITIVE DEFINITE. THIS COULD INDICATE A NEGATIVE VARIANCE/RESIDUAL VARIANCE FOR A LATENT VARIABLE, A CORRELATION GREATER OR EQUAL TO ONE BETWEEN TWO LATENT VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE THAN TWO LATENT VARIABLES.CHECK THE TECH4 OUTPUT FOR MORE INFORMATION. PROBLEM INVOLVING VARIABLE LIVWSMKR.

Advise?
 Linda K. Muthen posted on Monday, April 21, 2008 - 2:00 pm
If the variable referred to in the error message is an auxiliary variable, I think you can ignore the message.
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: