OpenMx Developer Forums

• 2010/10/03
  • Developers will now see submenus under the commit log
• 2009/10/11
  • Michael fixed Subscriptions.
  • Changed the title to "OpenMx" and hardcoded the h1 and title to included " - Advanced Structural Equation Modeling"
• 2009/10/05
  • Looks like Drupal 6.14 introduced a bug in forum replies. Fixed for now.
• 2009/10/03
  • Again, haven't done this for a while.

Perhaps this could be done by expanding the loading and saving of models to also include past fitting attempts. This way mxTryHard can pick up where left off when interrupted, and I can also page through the solutions of all the tries that mxTryHard made.

This should be a quick question considering it is probably common knowledge for all of you. I am looking for some clarification regarding the mxcompare output. Just to be certain, is "diffll" synonymous with chi-squared? From the openmx user guide I cannot see where else a chi-square output would be located. If this is not the case, how can one determine the chi-square value from the usual mxcompare output?

Thank you in advance and please excuse my extreme naivety.

Thanks for all the work on OpenMx!

I know about mxSave and mxRestore, but when I run mxRestore I still have to use mxRun to compute all the numerical derivs before I can print a summary() of the model's results. This can take a while if I am reading in many models.

Would it be possible for mxSave and mxRestore to also save/restore the derivs that are computed in the last step? This way I can run a model, save it, and then instantly load it later when I want to work with the results.

Thanks for your time!
Kyle

First, I would like to thank you for all your fantastic work.

I am wondering if the mxBoostrap() could return an object of class boot from the base boot package. A similar facility has been developed in rms::bootcov() function. This capability makes it standard on how to compute the confidence interval using different approaches available in the boot package.

http://openmx.psyc.virginia.edu/docs/OpenMx/2.6.7/_static/demo/OneFactorJoint_PathRaw.R

When I ran it, there were no estimates for the factor loadings.

This code doesn't include "facLoads". Shouldn't it?

oneFactorJointModel <- mxModel("Common Factor Model Path Specification", type="RAM",
manifestVars=c("x1","x2","x3","z1","z2","z3"), latentVars="F1",
dataRaw, resVars, latVar, means, thresholds)

I think this is what is needed:

I am currently trying to profile the mxRun function on OS X. I am stuck because the fortran code is not at the correct place in the call tree. I described the problem in detail here: http://stackoverflow.com/questions/32378821/profile-compiled-r-code-including-c-and-fortan-on-os-x. Does anybody have any experience with stuff like this?