randomDAG

After simulating a random topological order first (the order can also be prespecified), the algorithm connects two nodes with the porbability
probConnect.

 randomDAG 

 causal (topological) order 

The code computes the structural intervention distance (SID) between a true directed acyclic graph (DAG) and an estimated DAG. Definition and details about the implementation can be found in J. Peters and P. Bühlmann: "Structural intervention distance (SID) for evaluating causal graphs", Neural Computation 27, pages 771-799, 2015 <doi:10.1162/NECO_a_00708>.

Fred Gruber

Structural Intervention Distance

Jonas Peters

randomDAG function

<dl> <dt>p</dt>
<dd>number of nodes.</dd> <dt>probConnect</dt>
<dd>probability of connecting two nodes, determines the sparsity of the graph. Choosing probConnect = 2/(p-1), for example, leads to an expected number of p nodes.</dd> <dt>causalOrder</dt>
<dd>OPTIONAL: causal or topological order of the nodes. If not provided, the topological order is chosen randomly.</dd></dl>

Arguments

Jonas Peters &lt;jonas.peters@tuebingen.mpg.de&gt;

Author

outputs the adjacency matrix of a randomly generated directed acyclic graph (DAG). — randomDAG

<dl>

 <dt>p</dt>
<dd>number of nodes.</dd>

 <dt>probConnect</dt>
<dd>probability of connecting two nodes, determines the sparsity of the graph. Choosing probConnect = 2/(p-1), for example, leads to an expected number of p nodes.</dd>

 <dt>causalOrder</dt>
<dd>OPTIONAL: causal or topological order of the nodes. If not provided, the topological order is chosen randomly.</dd>

</dl>

randomDAG: outputs the adjacency matrix of a randomly generated directed acyclic graph (DAG).

Description

Usage

Value

Arguments

Author

See Also

Examples