Learning extremal graphical structures in high dimensions
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2021
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914134347481088 |
|---|---|
| author | Engelke, Sebastian Lalancette, Michaël Volgushev, Stanislav |
| author_facet | Engelke, Sebastian Lalancette, Michaël Volgushev, Stanislav |
| contents | Extremal graphical models encode the conditional independence structure of multivariate extremes. Key statistics for learning extremal graphical structures are empirical extremal variograms, for which we prove non-asymptotic concentration bounds that hold under general domain of attraction conditions. For the popular class of Hüsler--Reiss models, we propose a majority voting algorithm for learning the underlying graph from data through $L^1$ regularized optimization. Our concentration bounds are used to derive explicit conditions that ensure the consistent recovery of any connected graph. The methodology is illustrated through a simulation study as well as the analysis of river discharge and currency exchange data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2111_00840 |
| institution | arXiv |
| publishDate | 2021 |
| record_format | arxiv |
| spellingShingle | Learning extremal graphical structures in high dimensions Engelke, Sebastian Lalancette, Michaël Volgushev, Stanislav Statistics Theory 60G70, 62H22 (Primary) 62G30 (Secondary) Extremal graphical models encode the conditional independence structure of multivariate extremes. Key statistics for learning extremal graphical structures are empirical extremal variograms, for which we prove non-asymptotic concentration bounds that hold under general domain of attraction conditions. For the popular class of Hüsler--Reiss models, we propose a majority voting algorithm for learning the underlying graph from data through $L^1$ regularized optimization. Our concentration bounds are used to derive explicit conditions that ensure the consistent recovery of any connected graph. The methodology is illustrated through a simulation study as well as the analysis of river discharge and currency exchange data. |
| title | Learning extremal graphical structures in high dimensions |
| topic | Statistics Theory 60G70, 62H22 (Primary) 62G30 (Secondary) |
| url | https://arxiv.org/abs/2111.00840 |