Individually varying times PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Graciela posted on Tuesday, August 09, 2005 - 1:10 pm
In the study i am working on we have the following variables: education, sex, cohort1-cohort4 and score1-score5.
I am trying to run the following pgm:

VARIABLE: NAMES ARE ed sx age coh coh1-coh4 s1-s5;
CLASSES=c(1);
USEVARIABLES ARE ed,sx,coh1-coh4,y1-y5,x1-x5;
TSCORES=x1-x5;

DEFINE: y1=s1/10;y2=s2/10; y3=s3/10; y4=s4/10;y5=s5/10; x1=age;
x2=age+2;x3=age+7;x4=age+9;x5=age+13;
ANALYSIS: TYPE=MIXTURE RANDOM ;
STARTS=200 20;
algorithm=integration;
MODEL:
%OVERALL%
i s |y1-y5 AT x1-x5;
i ON sx ed coh1-coh3;
s ON sx ;
i*2.10 s*0.002;
OUTPUT: sampstat tech1 ;
SAVEDATA: FILE IS H:\simulationsmplus\dostext.txt;
RESULTS ARE H:\simulationsmplus\dos.dat;
SAVE=CPROBABILITIES;
SAVE=FSCORES;

But I have not succeeded yet in making it work.
What's wrong with it? I ran a similar program wher instead of x1-x5 I used times 0,2,7,9,13 that were the actual times from the first interview; but my interest is to run a model based on age rathar than times.
I would appreciatte any suggestions.
Best,
Graciela
 Linda K. Muthen posted on Tuesday, August 09, 2005 - 4:23 pm
Please send you input, output, data, 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: