Released on CRAN: March 23, 2023

The goric function has a new argument missing = and can be set to 'two-stage' (Savalei and Bentler, 2009). In this approach, missing data are imputed using an EM algorithm for continuous outcome variable. In addition, using the new argument auxiliary =, variables can be included to improve the imputation model.

The goric function can now handle objects of class restriktor, lm, numeric (vector with the parameter estimates of any model), lavaan, CTmeta, and rma.

The goric summary now shows the ratio of penalty weights and the ratio of likelihood weights.

The constraint syntax now knows how to handle constraints of format "(x1,x2,x3) > 0".