reproduce_paper_results/Readme.md

In nicolagnecco/causalXtreme: Causal discovery in heavy-tailed models

Readme

This folder contains the files to replicate the simulations and real-world experiments of Gnecco et al. (2019, https://arxiv.org/abs/1908.05097).

A brief introduction to the causalXtreme package can be found at https://nicolagnecco.github.io/causalXtreme.

The required R packages are: causalXtreme, tidyverse, tidyquant, doParallel, doRNG, rngtools, tictoc, latex2exp, tools, kableExtra, cowplot, evd, evir, Hmisc, timetk, tsibble, reshape2, ismev, and pracma.

This folder contains the following files:

• main.R, which runs all the R files to produce the results and figures of Sections 5.1–5.3
• simulation_0.R, …, simulation_3.R, which reproduce results of Section 5.1
• simulation_functions.R, which contains all the helper functions for results of Section 5.1
• produce_charts.R, which reproduces all figures of Section 5.1
• financial_data.R, which reproduces all results and figures of Section 5.2
• river_map.R, which reproduces the river map of Section 5.3
• river_data.R, which reproduces all remaining results and figures of Section 5.3

It also contains the following folders:

• data, which contains the raw data for the financial and river data experiments
• original_output, which contains the output of simulations of Section 5.1. These simulations can be reproduced using main.R, but they are computationally intensive
• output, which is an empty folder for the results produced by main.R

Warning: in order to run main.R set the working directory to /reproduce_paper_results.

References

Gnecco, Nicola, Nicolai Meinshausen, Jonas Peters, and Sebastian Engelke. 2019. “Causal Discovery in Heavy-Tailed Models.” *arXiv Preprint arXiv:1908.05097*.

nicolagnecco/causalXtreme documentation built on April 21, 2024, 4:22 a.m.