Estimating a Directed Tree for Extremes
Machine Learning
2023-12-29 v4 Machine Learning
Applications
Methodology
Abstract
We propose a new method to estimate a root-directed spanning tree from extreme data. A prominent example is a river network, to be discovered from extreme flow measured at a set of stations. Our new algorithm utilizes qualitative aspects of a max-linear Bayesian network, which has been designed for modelling causality in extremes. The algorithm estimates bivariate scores and returns a root-directed spanning tree. It performs extremely well on benchmark data and new data. We prove that the new estimator is consistent under a max-linear Bayesian network model with noise. We also assess its strengths and limitations in a small simulation study.
Cite
@article{arxiv.2102.06197,
title = {Estimating a Directed Tree for Extremes},
author = {Ngoc Mai Tran and Johannes Buck and Claudia Klüppelberg},
journal= {arXiv preprint arXiv:2102.06197},
year = {2023}
}
Comments
Extensive Revision. 54 pages, 26 Figures