Attachment | Size |
---|---|
Scatter plot of the MZ twin data [6] | 16.3 KB |
Scatter plot of the DZ twin data [7] | 11.98 KB |
Script to run ACE and saturated models [8] | 6.99 KB |
Hello,
I'm trying to run an ACE model for some imaging-derived measures and hope you could help me to solve some of the issues.
Initially, I am aiming to run the ACE model and test it against the saturated model to evaluate if fitting a twin model is appropriate for the data. In the ACE model I used sex and age as covariates, however, I'm struggling to incorporate those covariates in the saturated model. So the main question is how to appropriately incorporate covariates in the saturated model? (please find my script attached (data used is as presented in scatter plots - 60 DZ and 117 MZ pairs)).
In addition, when running the ACE model I quite often get a warning message 5 or 6, e.g.
"In model 'twinACE' Optimizer returned a non-zero status code 5. The Hessian at the solution does not appear to be convex. See ?mxCheckIdentification for possible diagnosis (Mx status RED)."
I tried to re-run the model twice by inputting the output of one run to the next run to resolve this issue, but it helped only in a few cases.
Using this data presented in the scatterplot it gives code 5 warning when fitting the saturated model. Is it likely that the data is problematic? or I'm running the model incorrectly?
I would very much appreciate your help.
Best regards,
Aurina