The revisions let you track differences between multiple versions of a post.
I am not sure if I have come across a minor bug in the OpenMx RAM specification. For some reason I cannot free up an element of the S matrix when fitting a twin sub-model. I have tried several different syntax variations (i.e., using the label name and specifically freeing the S parameter element) to free up the parameter, but for some reason the specified covariance parameter element is not being estimated and included in the OpenMx output summary. The odd thing about this is that when checking the S matrix specification, the parameter is set to TRUE. Is this a known bug?
This is not the appropriate spot for this question. Please make a forum post about it instead. I will be deleting this comment soon.
To address your question, that is definitely not the desired behavior. Could you provide a minimal code example that produces the bug? Or some part of the script that does this? It's difficult to track down where the problem is without some example to work with.
Another thing to check for is re-setting a parameter. If you set a parameter to free=TRUE in line 15 of your script and set it to free=FALSE in line 33, then it will silently be free=FALSE even though you set it to free=TRUE initially.
good if this page had a link to create new pages...
how to create a new page with link
seems not good that all the links are gone from the page?