Causal discovery in heavy-tailed models
From MaRDI portal
Publication:820829
DOI10.1214/20-AOS2021zbMATH Open1475.62130arXiv1908.05097OpenAlexW3187292592MaRDI QIDQ820829FDOQ820829
Authors: Nicola Gnecco, Nicolai Meinshausen, Jonas Peters, Sebastian Engelke 
Publication date: 28 September 2021
Published in: The Annals of Statistics (Search for Journal in Brave)
Abstract: Causal questions are omnipresent in many scientific problems. While much progress has been made in the analysis of causal relationships between random variables, these methods are not well suited if the causal mechanisms only manifest themselves in extremes. This work aims to connect the two fields of causal inference and extreme value theory. We define the causal tail coefficient that captures asymmetries in the extremal dependence of two random variables. In the population case, the causal tail coefficient is shown to reveal the causal structure if the distribution follows a linear structural causal model. This holds even in the presence of latent common causes that have the same tail index as the observed variables. Based on a consistent estimator of the causal tail coefficient, we propose a computationally highly efficient algorithm that estimates the causal structure. We prove that our method consistently recovers the causal order and we compare it to other well-established and non-extremal approaches in causal discovery on synthetic and real data. The code is available as an open-access R package.
Full work available at URL: https://arxiv.org/abs/1908.05097
Recommendations
Nonparametric estimation (62G05) Nonparametric regression and quantile regression (62G08) Statistics of extreme values; tail inference (62G32)
Cites Work
- Causation, prediction, and search
- An introduction to statistical modeling of extreme values
- Independent component analysis, a new concept?
- Title not available (Why is that?)
- Causal Inference for Statistics, Social, and Biomedical Sciences
- Extreme value theory. An introduction.
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Causality. Models, reasoning, and inference
- Title not available (Why is that?)
- Extremes on river networks
- Structural Intervention Distance for Evaluating Causal Graphs
- A simple general approach to inference about the tail of a distribution
- Investigating Causal Relations by Econometric Models and Cross-spectral Methods
- Estimation of the Marginal Expected Shortfall: the Mean When a Related Variable is Extreme
- Title not available (Why is that?)
- Identifiability of Gaussian structural equation models with equal error variances
- Regularly varying multivariate time series
- Stable graphical models
- Title not available (Why is that?)
- A linear non-Gaussian acyclic model for causal discovery
- Three approaches to the quantitative definition of information*
- PC algorithm for Gaussian copula graphical models
- Title not available (Why is that?)
- Causal reasoning with ancestral graphs
- Independence properties of directed markov fields
- CAM: causal additive models, high-dimensional order search and penalized regression
- ParceLiNGAM: A Causal Ordering Method Robust Against Latent Confounders
- Max-linear models on directed acyclic graphs
- Multiple block sizes and overlapping blocks for multivariate time series extremes
- Identifiability and estimation of recursive max‐linear models
- Graphical Models for Extremes
- Title not available (Why is that?)
- Estimation of causal effects using linear non-Gaussian causal models with hidden variables
Cited In (12)
- Title not available (Why is that?)
- Total positivity in multivariate extremes
- Regular variation in Hilbert spaces and principal component analysis for functional extremes
- Causality in extremes of time series
- Causal modelling of heavy-tailed variables and confounders with application to river flow
- Causal Discovery via Reproducing Kernel Hilbert Space Embeddings
- New challenges in the interplay between finance and insurance. Abstracts from the workshop held October 1--6, 2023
- Estimation and Inference of Extremal Quantile Treatment Effects for Heavy-Tailed Distributions
- Partial Tail-Correlation Coefficient Applied to Extremal-Network Learning
- Modeling panels of extremes
- Max-linear graphical models with heavy-tailed factors on trees of transitive tournaments
- Computational causal discovery: Advantages and assumptions
Uses Software
This page was built for publication: Causal discovery in heavy-tailed models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q820829)
