Nudging state-space models for Bayesian filtering under misspecified dynamics

Nudging is a popular algorithmic strategy in numerical filtering to deal with the problem of inference in high-dimensional dynamical systems. We demonstrate in this paper that general nudging techniques can also tackle another crucial statistical problem in filtering, namely the misspecification of...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gonzalez, Fabian, Akyildiz, O. Deniz, Crisan, Dan, Miguez, Joaquin
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Gonzalez, Fabian
Akyildiz, O. Deniz
Crisan, Dan
Miguez, Joaquin
description Nudging is a popular algorithmic strategy in numerical filtering to deal with the problem of inference in high-dimensional dynamical systems. We demonstrate in this paper that general nudging techniques can also tackle another crucial statistical problem in filtering, namely the misspecification of the transition model. Specifically, we rely on the formulation of nudging as a general operation increasing the likelihood and prove analytically that, when applied carefully, nudging techniques implicitly define state-space models (SSMs) that have higher marginal likelihoods for a given (fixed) sequence of observations. This provides a theoretical justification of nudging techniques as data-informed algorithmic modifications of SSMs to obtain robust models under misspecified dynamics. To demonstrate the use of nudging, we provide numerical experiments on linear Gaussian SSMs and a stochastic Lorenz 63 model with misspecified dynamics and show that nudging offers a robust filtering strategy for these cases.
doi_str_mv 10.48550/arxiv.2411.00218
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2411_00218</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2411_00218</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2411_002183</originalsourceid><addsrcrecordid>eNqFjrEOgjAQQLs4GPUDnLwfAFuEhFmjcdHJnVzolVxCC-mBkb83EHent7zkPaX2Rqd5WRT6iPHD7zTLjUm1zky5Vo_naBsODciAAyXSY03gO0utgOsinHEiYQzguB0ozuYYLEXwLNJTzY7Jgp0Ceq5lq1YOW6Hdjxt1uF1fl3uyhKs-ssc4VfNAtQyc_htfcJw8Bg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Nudging state-space models for Bayesian filtering under misspecified dynamics</title><source>arXiv.org</source><creator>Gonzalez, Fabian ; Akyildiz, O. Deniz ; Crisan, Dan ; Miguez, Joaquin</creator><creatorcontrib>Gonzalez, Fabian ; Akyildiz, O. Deniz ; Crisan, Dan ; Miguez, Joaquin</creatorcontrib><description>Nudging is a popular algorithmic strategy in numerical filtering to deal with the problem of inference in high-dimensional dynamical systems. We demonstrate in this paper that general nudging techniques can also tackle another crucial statistical problem in filtering, namely the misspecification of the transition model. Specifically, we rely on the formulation of nudging as a general operation increasing the likelihood and prove analytically that, when applied carefully, nudging techniques implicitly define state-space models (SSMs) that have higher marginal likelihoods for a given (fixed) sequence of observations. This provides a theoretical justification of nudging techniques as data-informed algorithmic modifications of SSMs to obtain robust models under misspecified dynamics. To demonstrate the use of nudging, we provide numerical experiments on linear Gaussian SSMs and a stochastic Lorenz 63 model with misspecified dynamics and show that nudging offers a robust filtering strategy for these cases.</description><identifier>DOI: 10.48550/arxiv.2411.00218</identifier><language>eng</language><subject>Mathematics - Probability ; Statistics - Computation</subject><creationdate>2024-10</creationdate><rights>http://creativecommons.org/licenses/by-sa/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2411.00218$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2411.00218$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Gonzalez, Fabian</creatorcontrib><creatorcontrib>Akyildiz, O. Deniz</creatorcontrib><creatorcontrib>Crisan, Dan</creatorcontrib><creatorcontrib>Miguez, Joaquin</creatorcontrib><title>Nudging state-space models for Bayesian filtering under misspecified dynamics</title><description>Nudging is a popular algorithmic strategy in numerical filtering to deal with the problem of inference in high-dimensional dynamical systems. We demonstrate in this paper that general nudging techniques can also tackle another crucial statistical problem in filtering, namely the misspecification of the transition model. Specifically, we rely on the formulation of nudging as a general operation increasing the likelihood and prove analytically that, when applied carefully, nudging techniques implicitly define state-space models (SSMs) that have higher marginal likelihoods for a given (fixed) sequence of observations. This provides a theoretical justification of nudging techniques as data-informed algorithmic modifications of SSMs to obtain robust models under misspecified dynamics. To demonstrate the use of nudging, we provide numerical experiments on linear Gaussian SSMs and a stochastic Lorenz 63 model with misspecified dynamics and show that nudging offers a robust filtering strategy for these cases.</description><subject>Mathematics - Probability</subject><subject>Statistics - Computation</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNqFjrEOgjAQQLs4GPUDnLwfAFuEhFmjcdHJnVzolVxCC-mBkb83EHent7zkPaX2Rqd5WRT6iPHD7zTLjUm1zky5Vo_naBsODciAAyXSY03gO0utgOsinHEiYQzguB0ozuYYLEXwLNJTzY7Jgp0Ceq5lq1YOW6Hdjxt1uF1fl3uyhKs-ssc4VfNAtQyc_htfcJw8Bg</recordid><startdate>20241031</startdate><enddate>20241031</enddate><creator>Gonzalez, Fabian</creator><creator>Akyildiz, O. Deniz</creator><creator>Crisan, Dan</creator><creator>Miguez, Joaquin</creator><scope>AKZ</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20241031</creationdate><title>Nudging state-space models for Bayesian filtering under misspecified dynamics</title><author>Gonzalez, Fabian ; Akyildiz, O. Deniz ; Crisan, Dan ; Miguez, Joaquin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2411_002183</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Mathematics - Probability</topic><topic>Statistics - Computation</topic><toplevel>online_resources</toplevel><creatorcontrib>Gonzalez, Fabian</creatorcontrib><creatorcontrib>Akyildiz, O. Deniz</creatorcontrib><creatorcontrib>Crisan, Dan</creatorcontrib><creatorcontrib>Miguez, Joaquin</creatorcontrib><collection>arXiv Mathematics</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Gonzalez, Fabian</au><au>Akyildiz, O. Deniz</au><au>Crisan, Dan</au><au>Miguez, Joaquin</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Nudging state-space models for Bayesian filtering under misspecified dynamics</atitle><date>2024-10-31</date><risdate>2024</risdate><abstract>Nudging is a popular algorithmic strategy in numerical filtering to deal with the problem of inference in high-dimensional dynamical systems. We demonstrate in this paper that general nudging techniques can also tackle another crucial statistical problem in filtering, namely the misspecification of the transition model. Specifically, we rely on the formulation of nudging as a general operation increasing the likelihood and prove analytically that, when applied carefully, nudging techniques implicitly define state-space models (SSMs) that have higher marginal likelihoods for a given (fixed) sequence of observations. This provides a theoretical justification of nudging techniques as data-informed algorithmic modifications of SSMs to obtain robust models under misspecified dynamics. To demonstrate the use of nudging, we provide numerical experiments on linear Gaussian SSMs and a stochastic Lorenz 63 model with misspecified dynamics and show that nudging offers a robust filtering strategy for these cases.</abstract><doi>10.48550/arxiv.2411.00218</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2411.00218
ispartof
issn
language eng
recordid cdi_arxiv_primary_2411_00218
source arXiv.org
subjects Mathematics - Probability
Statistics - Computation
title Nudging state-space models for Bayesian filtering under misspecified dynamics
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-26T05%3A53%3A30IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Nudging%20state-space%20models%20for%20Bayesian%20filtering%20under%20misspecified%20dynamics&rft.au=Gonzalez,%20Fabian&rft.date=2024-10-31&rft_id=info:doi/10.48550/arxiv.2411.00218&rft_dat=%3Carxiv_GOX%3E2411_00218%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true