A new version of the stagedtrees package (v2.3.0) is available on CRAN, with new structural learning algorithms for simple and context-specific staged trees, random generation of staged trees and faster internal computations.

NEWS

• stages_simplebhc searches in the space of simple staged trees, by iterative joining of positions.
• stages_csbhc learns stages by iteratively adding context-specific independences. The functions ci_matrices and join_all have also been added.
• sevt_simplify returns the simplified staged tree, where positions and stages are equivalent.
• random_sevt and random_parentslist generate random staged trees.
• depsubtree builds the dependency subtree.
• new functions to translate sevt and ceg objects to graphs (edge lists and igraph).
• sevt_fit earns the scope argument for partial re-fitting of staged trees, and the compute_logLik argument.
• stages_hclust can search the best cut of the hierarchical clustering with respect to a score function.
• new return value of stages and new indexing of stages via the class sevt.stgs.
• errors and warnings now use the cli package.
• faster internal functions (up to 10x speedup), more checks and better test coverage.