Regression modeling of multivariate precipitation extremes under regular variation
Rishikesh Yadav, Arnab Hazra · 2026
Read Regression modeling of multivariate precipitation extremes under regular variation by Rishikesh Yadav, Arnab Hazra (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-08865.txt.
Description
Heavy rain at its rarest is hard to predict, and Rishikesh Yadav and one co-author offer a lean method for it. Their entry in the EVA2025 data challenge models extreme rainfall from many runs of the CESM2 climate model. First, target levels are estimated at lower thresholds, then a simple regression grounded in regular-variation theory stretches them to very high quantiles.
Seasonal swings show up clearly, but no strong long-term trend appears across years, and runs can be treated as independent. Confidence bands come from a block bootstrap. The straightforward two-stage approach earned a joint second place in the challenge, trading some accuracy for far less computing than full likelihood methods.
Catalog record
| CREATOR | Rishikesh Yadav, Arnab Hazra |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-08865