Causal modelling of heavy-tailed variables and confounders with application to river flow

DOI10.1007/S10687-022-00456-4arXiv2110.06686OpenAlexW3206340597MaRDI QIDQ6176331FDOQ6176331

Olivier C. Pasche, V. Chavez-Demoulin, A. C. Davison

Publication date: 22 August 2023

Published in: Extremes (Search for Journal in Brave)

Abstract: Confounding variables are a recurrent challenge for causal discovery and inference. In many situations, complex causal mechanisms only manifest themselves in extreme events, or take simpler forms in the extremes. Stimulated by data on extreme river flows and precipitation, we introduce a new causal discovery methodology for heavy-tailed variables that allows the effect of a known potential confounder to be almost entirely removed when the variables have comparable tails, and also decreases it sufficiently to enable correct causal inference when the confounder has a heavier tail. We also introduce a new parametric estimator for the existing causal tail coefficient and a permutation test. Simulations show that the methods work well and the ideas are applied to the motivating dataset.

Full work available at URL: https://arxiv.org/abs/2110.06686

zbMATH Keywords

generalized Pareto distributioncausationextreme value statisticsconfoundercausal tail coefficient

Mathematics Subject Classification ID

Statistical distribution theory (62Exx)

Cites Work

• Causation, prediction, and search
• An introduction to statistical modeling of extreme values
• Causal inference using invariant prediction: identification and confidence intervals
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• PC algorithm for Gaussian copula graphical models
• Max-linear models on directed acyclic graphs
• Causal discovery in heavy-tailed models
• Climate extreme event attribution using multivariate peaks-over-thresholds modeling and counterfactual theory
• Estimating an extreme Bayesian network via scalings

Cited In (1)

• Causality in extremes of time series

   Recommendations

• A light-tailed conditionally heteroscedastic model with applications to river flows 👍 👎
• Causal discovery in heavy-tailed models 👍 👎
• Modeling and forecasting river flows by means of filtered Poisson processes 👍 👎
• Using filtered Poisson processes to model a river flow 👍 👎
• Non-negative time series models for dry river flow 👍 👎
• ON THE CAUSALITY TEST IN TIME SERIES MODELS WITH HEAVY-TAILED DISTRIBUTION 👍 👎
• Estimation of causal effects using linear non-Gaussian causal models with hidden variables 👍 👎

This page was built for publication: Causal modelling of heavy-tailed variables and confounders with application to river flow

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6176331)