2016 June ; 81(2): 535–549 | Michael C. Neale, Michael D. Hunter, Joshua Pritkin, Mahsa Zahery, Timothy R. Brick, Robert M. Kirkpatrick, Ryne Estabrook, Timothy C. Bates, Hermine H. Maes, and Steven M. Boker
The article introduces OpenMx 2.0, a software package for structural equation and other statistical modeling. Key features of OpenMx 2.0 include modular architecture, support for various optimization algorithms, and the ability to specify models using different methods such as LISREL syntax. The package now allows users to mix and match model expectations, fit functions, and optimizers, making it easier to fit complex models. New functionalities include Item Factor Analysis (IFA) and State-space modeling, which are implemented using open-source optimizers like CSOLNP. The article also highlights improvements in ease of use, such as helper functions for standardizing parameters and accessing model components through R S mechanisms. Additionally, OpenMx 2.0 supports parallel computation and provides detailed examples and scripts to illustrate its new features. The authors outline future developments, including plans for multilevel models and least squares fit functions.The article introduces OpenMx 2.0, a software package for structural equation and other statistical modeling. Key features of OpenMx 2.0 include modular architecture, support for various optimization algorithms, and the ability to specify models using different methods such as LISREL syntax. The package now allows users to mix and match model expectations, fit functions, and optimizers, making it easier to fit complex models. New functionalities include Item Factor Analysis (IFA) and State-space modeling, which are implemented using open-source optimizers like CSOLNP. The article also highlights improvements in ease of use, such as helper functions for standardizing parameters and accessing model components through R S mechanisms. Additionally, OpenMx 2.0 supports parallel computation and provides detailed examples and scripts to illustrate its new features. The authors outline future developments, including plans for multilevel models and least squares fit functions.