PAG2ADMG

Method to transform a PAG to its full Markov equivalent set of ADMGs

This repository contains complementary information for AAAI17 Submission Titled: PAG2ADMG: A Novel Methodology to Enumerate Causal Graph Structures.

There are two files in the 'Code' folder. One titled 'full_admg_learning.R' which has all methods of the algorithm, graph visualization etc. It uses two R packages 'ggm' and 'pcalg.' The second 'AAAI17_exp_script.R' runs a basic script to generate the two PAGs in the extended abstract as well as run the methods to be able to generate the full set of ADMGs off of those two PAGs.

There is one file in the 'Documents' folder. It is titled 'pag2admg-important-theorems.pdf' and is a pdf stating all five theorems from the extended abstract and giving proofs for each of them. This document provides the proof of correctness for the pag2admg method.

Packaging

Creation

install.packages("devtools") install.packages("roxygen2")

library(devtools) library(roxygen2)

devtools::create("PAG2ADMG")

How to load

devtools::install_github("HCSS-Data-Lab/PAG2ADMG")

Source

https://tinyheero.github.io/jekyll/update/2015/07/26/making-your-first-R-package.html