Genetic GxE model PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Chris Weber posted on Tuesday, June 29, 2010 - 3:48 pm
I cannot fit a common factor GxE model with twins data (similar to Harden et al- MPLUS/genetics section of website). I am getting troubling estimates. U, which is the estimate of c' is large and negative. At the maximum of the moderator c will be negative (?). Moderator ranges from 0 to 1.
f_t1 (v1); f_t2 (v2);

new(a c e t u);
v1=(a+t*covariate1)**2 + (c+u* covariate1)**2 +e**2;
v2=(a+t*covariate2)**2 + (c+u* covariate2)**2 +e**2;
mzc=(a+t* covariate1)*(a+t* covariate2) + (c+u* covariate1)*(c+u* covariate2);
dzc=0.50*(a+t* covariate1)*(a+t* covariate2) + (c+u* covariate1)*(c+u*covariate2);
F_T1 ON
COVARIATE1 0.737 0.169 4.348 0.000
F_T2 ON
COVARIATE2 0.737 0.169 4.348 0.000
New/Additional Parameters
A -0.467 0.305 -1.532 0.125
C 0.731 0.217 3.370 0.001
E 0.615 0.039 15.747 0.000
T -0.116 0.441 -0.264 0.792
U -1.081 0.293 -3.690 0.000
Thanks!
 Linda K. Muthen posted on Wednesday, June 30, 2010 - 4:47 pm
Please send your input, data, output, and license number to support@statmodel.com.
 Sophie van der Sluis posted on Tuesday, February 01, 2011 - 8:24 pm
Hi
I try to fit a bivariate GxE model in Mplus, where the moderator not only features as a variable whose variance is decomposed but also as a contraint variable.

I get the message:
*** FATAL ERROR
THIS PARAMETER RESTRICTION MODEL IS NOT AVAILABLE.

i.e., this model cannot be fit in Mplus?
thanks
Sophie
 Linda K. Muthen posted on Tuesday, February 01, 2011 - 9:26 pm
Please send your output and license number to support@statmodel.com.
 Erika Wolf posted on Tuesday, September 20, 2011 - 3:50 pm
I am trying to run a G by measured E ACE model. I get an error message that the computation could not be completed and that the correlation and residual correlation between A2 and A1 is 1.0.

I tried ALGORITHM=EM AND set MCONVERGENCE to .01 but this did not help. Please advise.

Model:
[x x2] (1);
x@0; x2@0;
[z z2] (11);
z (13); z2 (13);


A1 BY x*.64 (3); A2 BY x2*.64 (3);
C1 BY x@0 (4); C2 BY x2@0 (4);
E1 BY x*.76 (5); E2 BY x2*.76 (5);

[A1-E2@0];
A1-E2@1;

X on z (12); X2 on z2(12);

GXz |A1 XWITH z;
GXz2|A2 XWITH z2;

x on GXz (14); x2 ON GXz2 (14);

a1 with a2@1;
C1 WITH C2@1;
e1 with e2@0;
[I'm omitting the other ACE correlations @0 here to save space]

MODEL DZ:
A1 WITH A2@.5;
 Linda K. Muthen posted on Tuesday, September 20, 2011 - 4:27 pm
Please send the output and your 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: