Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Robust statistical jump models

Lindström, Erik LU orcid and Persson, Jonatan LU (2025) Forecasting Financial Markets Conference
Abstract (Swedish)
Statistical Jump Models are a class of regime identification models,
similar to Hidden Markov Models, with the added advantage of facilitating
large-scale integration of exogenous variables that influence the
latent regime-switching process. Recent advancements in SJMs include
the introduction of automatic feature selection techniques, which eliminate
irrelevant exogenous information and enhance the model’s ability to
identify regimes more effectively.
We generalize the regular Statistical Jump Model by considering robust
distances measures, while preserving the feature selection. This is
important for financial data, as these are often heavy-tailed.
The simulation study confirm that the proposed... (More)
Statistical Jump Models are a class of regime identification models,
similar to Hidden Markov Models, with the added advantage of facilitating
large-scale integration of exogenous variables that influence the
latent regime-switching process. Recent advancements in SJMs include
the introduction of automatic feature selection techniques, which eliminate
irrelevant exogenous information and enhance the model’s ability to
identify regimes more effectively.
We generalize the regular Statistical Jump Model by considering robust
distances measures, while preserving the feature selection. This is
important for financial data, as these are often heavy-tailed.
The simulation study confirm that the proposed modified algorithm
is outperforming the regular statistical jump model on heavy-tailed data
but not on light tailed data, as predicted by statistical theory. We also
find that the corresponding latent regime process is more persistent when
estimated on index returns than comparable estimates derived using Hidden
Markov Models or the regular statistical jump model, an obvious
advantage for any trading strategy based on the latent regime process. (Less)
Please use this url to cite or link to this publication:
author
and
organization
publishing date
type
Contribution to conference
publication status
published
subject
conference name
Forecasting Financial Markets Conference
conference location
Venice, Italy
conference dates
2025-05-21 - 2025-05-23
language
English
LU publication?
yes
id
82a26049-a54e-4aba-99c6-5f685e18c7ba
date added to LUP
2025-06-05 16:38:59
date last changed
2025-08-12 16:16:43
@misc{82a26049-a54e-4aba-99c6-5f685e18c7ba,
  abstract     = {{Statistical Jump Models are a class of regime identification models,<br/>similar to Hidden Markov Models, with the added advantage of facilitating<br/>large-scale integration of exogenous variables that influence the<br/>latent regime-switching process. Recent advancements in SJMs include<br/>the introduction of automatic feature selection techniques, which eliminate<br/>irrelevant exogenous information and enhance the model’s ability to<br/>identify regimes more effectively.<br/>We generalize the regular Statistical Jump Model by considering robust<br/>distances measures, while preserving the feature selection. This is<br/>important for financial data, as these are often heavy-tailed.<br/>The simulation study confirm that the proposed modified algorithm<br/>is outperforming the regular statistical jump model on heavy-tailed data<br/>but not on light tailed data, as predicted by statistical theory. We also<br/>find that the corresponding latent regime process is more persistent when<br/>estimated on index returns than comparable estimates derived using Hidden<br/>Markov Models or the regular statistical jump model, an obvious<br/>advantage for any trading strategy based on the latent regime process.}},
  author       = {{Lindström, Erik and Persson, Jonatan}},
  language     = {{eng}},
  month        = {{05}},
  title        = {{Robust statistical jump models}},
  year         = {{2025}},
}