Second OpenMx 2.0 beta is released

AdminRobK's picture

The second beta of OpenMx version 2.0 is now publicly available! Interested users can download and install this new version by copy-pasting source('http://openmx.psyc.virginia.edu/getOpenMxBeta.R') into the R command line and pressing 'Return'. Documentation for the beta is available here.

This second beta incorporates a number of minor adjustments and bug fixes that make it an improvement over the first beta. Especially notable new features include the following:

  • CSOLNP, a new, open-source numerical optimizer written in C++, was introduced in the first beta. Now, most known CSOLNP bugs have been repaired. For certain analyses, CSOLNP can now be faster than NPSOL, and we are aware of several cases where it succeeds where NPSOL fails!
  • The first beta introduced the split between a model's expectation (how it's specified) and its fit (its objective function to be minimized). It is now easier to use Mx Algebras in the new expectation functions.
  • The printed summary() output of an MxModel is now more concise (by default); some of it has been made easier to read.
  • Item Factor Analysis, also known as Item Response Theory. That's right--the versatility of OpenMx has now been extended to IRT!

Users should be aware of the following issues:

  • IMPORTANT: Windows users should NOT try to run OpenMx with multiple threads. Doing so could crash R.
  • The log of changes in the beta version's User Guide is not current.
  • Users should get into the habit of ALWAYS using the $ accessor instead of the @ accessor. For example, myModelRun@output$estimate would now be myModelRun$output$estimate. We CANNOT guarantee that every usage of the @ accessor that worked with versions 1.3/1.4 will continue to work in 2.0.
  • As of this beta, CSOLNP still had a known issue pertaining to matrix conformability.

    We are eager to receive user feedback about this beta of OpenMx 2.0! Please post questions, comments, and issues to the OpenMx Forums.