ESG narrative analysis and prediction of financial index volatility using text mining and machine learning

Autores
Mercer, Antonio Carlos; Filho, Zacarias Curi; Póvoa, Ângela Cristiane Santos
Año de publicación
2025
Idioma
inglés
Tipo de recurso
documento de conferencia
Estado
versión aceptada
Descripción
Motivated by the growing significance of Environmental, Social, and Governance (ESG) factors and the market's pronounced sensitivity to real-time information, this study empirically examines the reaction of the New York Stock Exchange (NYSE) to ESG-related news. Leveraging a specialized news corpus, our methodology integrates a Large Language Model (LLM)-driven Natural Language Processing (NLP) pipeline with a classic event study to assess abnormal returns and trading volume. Our findings indicate a pronounced asymmetry in market reaction, whereby the financial penalties associated with negative news substantially exceed the rewards from positive news. Furthermore, we identify a selective and heterogeneous response: the market distinguishes between the distinct impacts of the Environmental, Social, and Governance pillars and focuses its reaction primarily on events of high materiality. Ultimately, this research establishes that the price discovery of ESG information is a dynamic process, conditional on news sentiment, thematic pillar, and the materiality of the reported event
Fil: Filho, Zacarias Curi. Pontifical Catholic University of Parana; Brasil.
Fil: Mercer, Antonio Carlos. Pontifical Catholic University of Parana; Brasil.
Fil: Póvoa, Ângela Cristiane Santos. Pontifical Catholic University of Parana; Brasil.
Materia
Artificial intelligence
Data processing
Data analysis
Economic models
Economic information
Speeches
Markets
Social responsibility
Inteligencia artificial
Procesamiento de datos
Análisis de datos
Modelo económico
Información económica
Discurso
Mercado
Responsabilidad social
Intelligence artificielle
Traitement des données
Analyse de données
Modèle économique
Information économique
Discours
Marché
Responsabilité sociale
Nivel de accesibilidad
acceso abierto
Condiciones de uso
http://creativecommons.org/licenses/by/4.0/
Repositorio
Repositorio Digital (UNVM)
Institución
Universidad Nacional de Villa María
OAI Identificador
oai:biblio.unvm.edu.ar:48195

id RDUNVM_67cf44becfef1b16ac0372a147b1b18d
oai_identifier_str oai:biblio.unvm.edu.ar:48195
network_acronym_str RDUNVM
repository_id_str 4256
network_name_str Repositorio Digital (UNVM)
spelling ESG narrative analysis and prediction of financial index volatility using text mining and machine learningMercer, Antonio CarlosFilho, Zacarias CuriPóvoa, Ângela Cristiane SantosArtificial intelligenceData processingData analysisEconomic modelsEconomic informationSpeechesMarketsSocial responsibilityInteligencia artificialProcesamiento de datosAnálisis de datosModelo económicoInformación económicaDiscursoMercadoResponsabilidad socialIntelligence artificielleTraitement des donnéesAnalyse de donnéesModèle économiqueInformation économiqueDiscoursMarchéResponsabilité socialeMotivated by the growing significance of Environmental, Social, and Governance (ESG) factors and the market's pronounced sensitivity to real-time information, this study empirically examines the reaction of the New York Stock Exchange (NYSE) to ESG-related news. Leveraging a specialized news corpus, our methodology integrates a Large Language Model (LLM)-driven Natural Language Processing (NLP) pipeline with a classic event study to assess abnormal returns and trading volume. Our findings indicate a pronounced asymmetry in market reaction, whereby the financial penalties associated with negative news substantially exceed the rewards from positive news. Furthermore, we identify a selective and heterogeneous response: the market distinguishes between the distinct impacts of the Environmental, Social, and Governance pillars and focuses its reaction primarily on events of high materiality. Ultimately, this research establishes that the price discovery of ESG information is a dynamic process, conditional on news sentiment, thematic pillar, and the materiality of the reported eventFil: Filho, Zacarias Curi. Pontifical Catholic University of Parana; Brasil.Fil: Mercer, Antonio Carlos. Pontifical Catholic University of Parana; Brasil.Fil: Póvoa, Ângela Cristiane Santos. Pontifical Catholic University of Parana; Brasil.2025info:eu-repo/semantics/conferenceObjectinfo:eu-repo/semantics/acceptedVersionhttp://purl.org/coar/resource_type/c_5794info:ar-repo/semantics/documentoDeConferenciaapplication/pdfhttp://biblio.unvm.edu.ar/opac_css/index.php?lvl=cmspage&pageid=9&id_notice=481954819520251112u u u0argy0103 baeng1020191Villa María (inhabited place)info:eu-repo/semantics/openAccesshttp://creativecommons.org/licenses/by/4.0/Creative Commons Attribution 4.0 International (CC BY 4.0)reponame:Repositorio Digital (UNVM)instname:Universidad Nacional de Villa María2026-11-05T11:53:23Zoai:biblio.unvm.edu.ar:48195instacron:UNVMInstitucionalhttp://biblioteca.unvm.edu.ar/Repositorio/index.htmlUniversidad públicaNo correspondehttp://biblio.unvm.edu.ar/ws/PMBOAI ggomez@unvm.edu.ar;gustavo0306@gmail.comArgentinaNo correspondeNo correspondeNo correspondeopendoar:42562026-11-05 11:53:24.279Repositorio Digital (UNVM) - Universidad Nacional de Villa Maríafalse
dc.title.none.fl_str_mv ESG narrative analysis and prediction of financial index volatility using text mining and machine learning
title ESG narrative analysis and prediction of financial index volatility using text mining and machine learning
spellingShingle ESG narrative analysis and prediction of financial index volatility using text mining and machine learning
Mercer, Antonio Carlos
Artificial intelligence
Data processing
Data analysis
Economic models
Economic information
Speeches
Markets
Social responsibility
Inteligencia artificial
Procesamiento de datos
Análisis de datos
Modelo económico
Información económica
Discurso
Mercado
Responsabilidad social
Intelligence artificielle
Traitement des données
Analyse de données
Modèle économique
Information économique
Discours
Marché
Responsabilité sociale
title_short ESG narrative analysis and prediction of financial index volatility using text mining and machine learning
title_full ESG narrative analysis and prediction of financial index volatility using text mining and machine learning
title_fullStr ESG narrative analysis and prediction of financial index volatility using text mining and machine learning
title_full_unstemmed ESG narrative analysis and prediction of financial index volatility using text mining and machine learning
title_sort ESG narrative analysis and prediction of financial index volatility using text mining and machine learning
dc.creator.none.fl_str_mv Mercer, Antonio Carlos
Filho, Zacarias Curi
Póvoa, Ângela Cristiane Santos
author Mercer, Antonio Carlos
author_facet Mercer, Antonio Carlos
Filho, Zacarias Curi
Póvoa, Ângela Cristiane Santos
author_role author
author2 Filho, Zacarias Curi
Póvoa, Ângela Cristiane Santos
author2_role author
author
dc.subject.none.fl_str_mv Artificial intelligence
Data processing
Data analysis
Economic models
Economic information
Speeches
Markets
Social responsibility
Inteligencia artificial
Procesamiento de datos
Análisis de datos
Modelo económico
Información económica
Discurso
Mercado
Responsabilidad social
Intelligence artificielle
Traitement des données
Analyse de données
Modèle économique
Information économique
Discours
Marché
Responsabilité sociale
topic Artificial intelligence
Data processing
Data analysis
Economic models
Economic information
Speeches
Markets
Social responsibility
Inteligencia artificial
Procesamiento de datos
Análisis de datos
Modelo económico
Información económica
Discurso
Mercado
Responsabilidad social
Intelligence artificielle
Traitement des données
Analyse de données
Modèle économique
Information économique
Discours
Marché
Responsabilité sociale
dc.description.none.fl_txt_mv Motivated by the growing significance of Environmental, Social, and Governance (ESG) factors and the market's pronounced sensitivity to real-time information, this study empirically examines the reaction of the New York Stock Exchange (NYSE) to ESG-related news. Leveraging a specialized news corpus, our methodology integrates a Large Language Model (LLM)-driven Natural Language Processing (NLP) pipeline with a classic event study to assess abnormal returns and trading volume. Our findings indicate a pronounced asymmetry in market reaction, whereby the financial penalties associated with negative news substantially exceed the rewards from positive news. Furthermore, we identify a selective and heterogeneous response: the market distinguishes between the distinct impacts of the Environmental, Social, and Governance pillars and focuses its reaction primarily on events of high materiality. Ultimately, this research establishes that the price discovery of ESG information is a dynamic process, conditional on news sentiment, thematic pillar, and the materiality of the reported event
Fil: Filho, Zacarias Curi. Pontifical Catholic University of Parana; Brasil.
Fil: Mercer, Antonio Carlos. Pontifical Catholic University of Parana; Brasil.
Fil: Póvoa, Ângela Cristiane Santos. Pontifical Catholic University of Parana; Brasil.
description Motivated by the growing significance of Environmental, Social, and Governance (ESG) factors and the market's pronounced sensitivity to real-time information, this study empirically examines the reaction of the New York Stock Exchange (NYSE) to ESG-related news. Leveraging a specialized news corpus, our methodology integrates a Large Language Model (LLM)-driven Natural Language Processing (NLP) pipeline with a classic event study to assess abnormal returns and trading volume. Our findings indicate a pronounced asymmetry in market reaction, whereby the financial penalties associated with negative news substantially exceed the rewards from positive news. Furthermore, we identify a selective and heterogeneous response: the market distinguishes between the distinct impacts of the Environmental, Social, and Governance pillars and focuses its reaction primarily on events of high materiality. Ultimately, this research establishes that the price discovery of ESG information is a dynamic process, conditional on news sentiment, thematic pillar, and the materiality of the reported event
publishDate 2025
dc.date.none.fl_str_mv 2025
dc.type.none.fl_str_mv info:eu-repo/semantics/conferenceObject
info:eu-repo/semantics/acceptedVersion
http://purl.org/coar/resource_type/c_5794
info:ar-repo/semantics/documentoDeConferencia
format conferenceObject
status_str acceptedVersion
dc.identifier.none.fl_str_mv http://biblio.unvm.edu.ar/opac_css/index.php?lvl=cmspage&pageid=9&id_notice=48195
48195
20251112u u u0argy0103 ba
url http://biblio.unvm.edu.ar/opac_css/index.php?lvl=cmspage&pageid=9&id_notice=48195
identifier_str_mv 48195
20251112u u u0argy0103 ba
dc.language.none.fl_str_mv eng
language eng
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
http://creativecommons.org/licenses/by/4.0/
Creative Commons Attribution 4.0 International (CC BY 4.0)
eu_rights_str_mv openAccess
rights_invalid_str_mv http://creativecommons.org/licenses/by/4.0/
Creative Commons Attribution 4.0 International (CC BY 4.0)
dc.format.none.fl_str_mv application/pdf
dc.coverage.none.fl_str_mv 1020191
Villa María (inhabited place)
dc.source.none.fl_str_mv reponame:Repositorio Digital (UNVM)
instname:Universidad Nacional de Villa María
reponame_str Repositorio Digital (UNVM)
collection Repositorio Digital (UNVM)
instname_str Universidad Nacional de Villa María
repository.name.fl_str_mv Repositorio Digital (UNVM) - Universidad Nacional de Villa María
repository.mail.fl_str_mv ggomez@unvm.edu.ar;gustavo0306@gmail.com
_version_ 1881032331961040896
score 12.823554