OpenMx 2.14 released!

We are pleased to announce the official release of OpenMx version 2.14.11. Click here for instructions on how to install the package from our repository. As usual, our repository has package binaries for Windows and MacOS, and source tarballs for Linux/GNU and other non-Mac Unix-likes, all of which come with the proprietary NPSOL optimizer. Alternately, users may install the fully open-source build of the new version from CRAN.

New Features, Performance Improvements, and Tweaks Since v2.13:

• A number of OpenMx error messages have been improved to be more helpful.
• It is now possible to compile NPSOL-enabled OpenMx under Linux/GNU with gcc version 8 or later.
• mxGenerateData() has a new argument, empirical, which if TRUE (FALSE is the default), generates a dataset having a covariance matrix exactly equal to the user-provided covariance matrix (a la MASS::mvrnorm()).
• mxGenerateData() can now return a covariance matrix if it is provided with an MxModel containing MxData of type="cov".
• The unfiltered expected covariance matrix returned in MxExpectationRAM now has dimnames.

Bug-fixes Since v2.13:

• Several serious bugs in mxPower() and mxPowerSearch(), which caused those functions not to respond correctly to their arguments, have been repaired. Users are advised to check results obtained from those two functions in previous versions of the package.
• A bug in mxFactorScores(), involving regression scores for RAM models, has been repaired. This bug would cause the latent-variable means to be calculated incorrectly in some cases.
• A bug in omxCheckWithinPercentError(), which caused the function to behave inappropriately if its first argument was negative, has been repaired.

Known issues

• summary() still counts redundant equality MxConstraints as though they were nonredundant (which is not a new issue, and likely has always existed in OpenMx). Therefore, it can get the model degrees-of-freedom wrong if redundant equality constraints are present in the model. Note that as of the v2.13.2 release, the package's NEWS.md file incorrectly implied that this issue was resolved.
• summary() reports only a few fit indices if the model is using the WLS fitfunction.
• The fit indices reported in summary() are wrong if the observed data are a correlation matrix (i.e., if type="cor" is passed to mxData().