This repository is part of the supplementary material for the paper:
N. Tagasovska, T. Vatter, V. Chavez-Demoulin, "Nonparametric Quantile-Based Causal Discovery”, 2018.
Availabe on arXiv: https://arxiv.org/pdf/1801.10579.pdf.

CONTENTS
———————————————————

qccd_code2018/
|-R/ (all of our R code lives here)
|---run_benchmarks/ (scripts for running qccd across different benchmarks)
|-data/ 
|—ANHNMN/ (simulated)
|-----AN-Gauss/
|-----AN-s/
|-----HN/
|-----HN-s/
|-----MN/
|-----MN-s/
|---SIM_pairs/ (Mooij 2016) 
|-----SIM/
|-----SIM-G/
|-----SIM-c/
|-----SIM-ln/
|---tuebingen_benchmark/ (real data Mooij 2016) 

Set working directory to setwd("~/qccd_code2018/R")
The file ~R/cd_utils.R contains a wrapper around our causal discovery method.

EXAMPLES
———————————————————
See R/run_benchmarks for examples.


PACKAGES REQUIRED
———————————————————
install.packages(“Hmisc”)
install.packages(“rvinecopulib”)
install.packages(“statmod”)
install.packages(“tnagler”)
install.packages(“igraph”)
install.packages(“pcalg”)
install.packages(“ggraph”)

COPYING
———————————————————
See COPYING for license terms.