PC (pcalg)
Module name |
|
Package |
|
Version |
2.7-8 |
Language |
|
Docs |
|
Paper |
Spirtes and Glymour[1] |
Graph type |
|
MCMC |
No |
Edge constraints |
|
Data type |
C, D |
Data missingness |
|
Intervention type |
|
Docker |
PC
The Peter and Clark (PC) algorithm Spirtes and Glymour[1], is a constraint based method consisting of two main steps. The first step is called the adjacency search and amounts to finding the undirected skeleton of the DAG. The second step amounts to estimating a CPDAG.
Some fields described
edgeConstraintsName of the JSON file containing background knowledge
Example JSON
[
{
"id": "pc-gaussCItest",
"alpha": [
0.001,
0.01,
0.05,
0.1
],
"NAdelete": true,
"mmax": "Inf",
"u2pd": "relaxed",
"skelmethod": "stable",
"conservative": false,
"majrule": false,
"solveconfl": false,
"numCores": 1,
"verbose": false,
"indepTest": "gaussCItest",
"timeout": null,
"edgeConstraints": "edgeConstraints.json"
},
{
"id": "pc-binCItest",
"alpha": [
0.01,
0.05,
0.1
],
"NAdelete": true,
"mmax": "Inf",
"u2pd": "relaxed",
"skelmethod": "stable",
"conservative": false,
"majrule": false,
"solveconfl": false,
"numCores": 1,
"verbose": false,
"indepTest": "binCItest",
"timeout": null,
"edgeConstraints": "edgeConstraints.json"
}
]