NEWS.md

In BCDAG: Bayesian Structure and Causal Learning of Gaussian Directed Graphs

BCDAG 1.1.3

• Corrected tests which were returning errors only on some flavors

BCDAG 1.1.2

• Improved documentation of learn_DAG()

BCDAG 1.1.1

New functions

• New confint() method for objects of class bcdagCE

BCDAG 1.1.0

New functions

• New print() and plot methods for objects of class bcdag;
• New object class bcdagCE, containing the output of function get_causaleffect(), and associated print(), summary() and plot methods;
• New function get_neighboringDAGs() to obtain all the DAGs that can be reached from a given DAG through an edge addition, removal or reversal;
• New function as_graphNEL() easily converting an adjacency matrix into a graphNEL object;

Minor improvements and fixes

• Faster matrix inversion in DW_nodelml() and internal function rnodeDAGWishart()via Cholesky decomposition;
• Faster matrix crossproduct via crossprod(X) in learn_DAG();
• get_diagnostics() now allows for different graphical settings and selection of specific nodes of the DAG;
• Bug correction in summary() method for object of class bcdag;

BCDAG 1.0

• First release of the package;

BCDAG documentation built on April 4, 2025, 1:41 a.m.