Flexible Tails for Normalising Flows, with Application to the Modelling of Financial Return Data

Tennessee Hickling, Dennis Prangle
11/1/2023
stat.MLcs.LG

Official code: Tennessee-Wallaceh/tailnflows (1 star, last commit 2025-07-28).

Abstract

We propose a transformation capable of altering the tail properties of a distribution, motivated by extreme value theory, which can be used as a layer in a normalizing flow to approximate multivariate heavy tailed distributions. We apply this approach to model financial returns, capturing potentially extreme shocks that arise in such data. The trained models can be used directly to generate new synthetic sets of potentially extreme returns

View on arXivView PDF

Code Implementations(1)

Tennessee-Wallaceh/tailnflowsOfficial100%

Repository for experiments related to "flexible tails for normalising flows"

12Python, Jupyter NotebookJun 5, 20231 years ago

Similar Papers

Flexible Tails for Normalizing Flows

84% match3 repos12024

Simulation of Multivariate Extremes: a Wasserstein-Aitchison GAN approach

80% match2025

A VAE Approach to Sample Multivariate Extremes

79% match2023

Phase-Type Variational Autoencoders for Heavy-Tailed Data

77% match2026

Classification of Heavy-tailed Features in High Dimensions: a Superstatistical Approach

75% match1 repo2023

On Regression in Extreme Regions

75% match1 repo12023

Cite this paper

@article{hickling2023flexible,
  title  = {Flexible Tails for Normalising Flows, with Application to the Modelling of Financial Return Data},
  author = {Tennessee Hickling and Dennis Prangle},
  year   = {2023},
  eprint = {2311.00580},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2311.00580v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2311.00580.svg)](https://codeofpaper.com/paper/2311.00580)

Discussion