OpenMx is a package for structural equation modeling, matrix algebra optimization and other statistical estimation problems.
OpenMx is a package for algebra optimization and statistical estimation problems using matrix algebra. The OpenMx library defines a set of S4 classes and functions used to create them. The majority of these classes are used as arguments in models, which may include data, matrices, algebras, bounds and constraints. These models are then paired with objective functions, either existing (maximum likelihood, FIML) or user-defined with included algebra functions. These models can then be optimized, resulting in parameter estimation, algebra evaluation, and output for additional models.
Objects used or created by OpenMx may be of the following classes: MxAlgebra, MxBounds, MxCI, MxConstraint, MxData, MxMatrix, MxModel, and MxPath. Objects of these classes may be created by the following OpenMx functions: mxAlgebra, mxBounds, mxCI, mxConstraint, mxData, mxMatrix, mxModel, and mxPath. The functions mxAlgebraObjective, mxFIMLObjective, mxMLObjective and mxRAMObjective create objective functions for model estimation. Models which include objective functions may be estimated using the mxRun function.
The OpenMx User's guide can be found at http://openmx.psyc.virginia.edu/documentation.