News sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock markets

Detalhes bibliográficos
Ano de defesa: 2025
Autor(a) principal: Cambraia, Nicolas Behar
Orientador(a): Pinto, Afonso de Campos, Botelho, Marcos Antonio
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: eng
Instituição de defesa: Não Informado pela instituição
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Previsão de volatilidade
Sentimento de notícias
Palavras-chave em Inglês:
HAR-X
Volatility forecasting
News sentiment
Link de acesso: https://hdl.handle.net/10438/37244
Resumo: The objective of this work is to apply to stocks traded on the Brazilian stock exchange the Heterogeneous Autoregressive model integrated with the use of news sentiment scores as an exogenous variable in order to estimate the effect of news on stock volatility. The volatility of the assets is obtained from the opening, minimum, maximum, and closing prices over one-hour intervals, using the Garman-Klass estimator. The sentiment scores are derived from a zero-shot classifier and proposed hypotheses, using as input approximately 2500 financial news from May 2023 to March 2025, with the intention of estimating the impact of news on stocks. Machine learning metrics such as R 2 and Mean Squared Error were compared between the models with and without exogenous variables for each evaluated asset. The inclusion of sentiment scores led to an overall improvement in performance. Furthermore, the analysis revealed a consistent lag structure across all evaluated stocks, characterized by similar linear coefficients associated with the lagged news sentiment variables. This pattern held irrespective of the industry in which each company operates, suggesting a generalized temporal response to news sentiment in volatility modeling.
id FGV_0b8217d7978c06df422e3781691bd36d
oai_identifier_str oai:repositorio.fgv.br:10438/37244
network_acronym_str FGV
network_name_str Repositório Institucional do FGV (FGV Repositório Digital)
repository_id_str
spelling Cambraia, Nicolas BeharEscolas::EESPAndrade, Douglas Coimbra dePinto, Afonso de CamposBotelho, Marcos Antonio2025-07-21T20:06:25Z2025-07-21T20:06:25Z2025-06-24https://hdl.handle.net/10438/37244The objective of this work is to apply to stocks traded on the Brazilian stock exchange the Heterogeneous Autoregressive model integrated with the use of news sentiment scores as an exogenous variable in order to estimate the effect of news on stock volatility. The volatility of the assets is obtained from the opening, minimum, maximum, and closing prices over one-hour intervals, using the Garman-Klass estimator. The sentiment scores are derived from a zero-shot classifier and proposed hypotheses, using as input approximately 2500 financial news from May 2023 to March 2025, with the intention of estimating the impact of news on stocks. Machine learning metrics such as R 2 and Mean Squared Error were compared between the models with and without exogenous variables for each evaluated asset. The inclusion of sentiment scores led to an overall improvement in performance. Furthermore, the analysis revealed a consistent lag structure across all evaluated stocks, characterized by similar linear coefficients associated with the lagged news sentiment variables. This pattern held irrespective of the industry in which each company operates, suggesting a generalized temporal response to news sentiment in volatility modeling.O objetivo deste trabalho e aplicar aos ativos negociados na bolsa de valores brasileira o modelo Heterogeneous Autorregressive (HAR) integrado com o uso de escores de sentimento extraídos de notícias como variável exógena, com o intuito de estimar o efeito das notícias sobre a volatilidade das ações. A volatilidade dos ativos é obtida a partir dos preços de abertura, mínima, máxima é fechamento em intervalos de uma hora, utilizando o estimador de Garman-Klass. Os escores de sentimento são derivados de um classificador zero-shot e hipótese propostas, com cerca de 2.500 notícias financeiras como entrada, abrangendo o período de maio de 2023 a março de 2025, visando estimar o impacto das notícias sobre os ativos. Métricas de aprendizado de máquina, como R 2 e erro quadrático médio (MSE), foram comparadas entre os modelos com e sem variáveis exógenas, para cada ativo avaliado. A inclusão dos escores de sentimento resultou em uma melhoria geral de desempenho. Ademais, a análise revelou uma estrutura de defasagens consistente entre todas as ações avaliadas, caracterizada por coeficientes lineares semelhantes associados as variáveis de sentimento de notícias. Esse padrão foi observado independentemente do setor de atuação das empresas, o que pode indicar uma resposta temporal generalizada ao ˜ sentimento de notícias no contexto da modelagem da volatilidade.engHAR-XVolatility forecastingNews sentimentPrevisão de volatilidadeSentimento de notíciasEconomiaMercado de capitais - BrasilMercado de ações - PrevisãoVolatilidade (Finanças)Processo decisórioNews sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock marketsinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisinfo:eu-repo/semantics/openAccessreponame:Repositório Institucional do FGV (FGV Repositório Digital)instname:Fundação Getulio Vargas (FGV)instacron:FGVORIGINALDissertação_Mestrado_FGV_Nicolas_Behar_Cambraia.pdfDissertação_Mestrado_FGV_Nicolas_Behar_Cambraia.pdfPDFapplication/pdf8369419https://repositorio.fgv.br/bitstreams/7b46cf49-1723-44b1-9bb5-0bf3a2b6cdf0/download6d61b7437ac819b9e744a1a3f174f875MD51trueAnonymousREADLICENSElicense.txtlicense.txttext/plain; charset=utf-85112https://repositorio.fgv.br/bitstreams/84d2ac30-17e1-4fab-9333-ef37950678e6/download2a4b67231f701c416a809246e7a10077MD52falseAnonymousREADTEXTDissertação_Mestrado_FGV_Nicolas_Behar_Cambraia.pdf.txtDissertação_Mestrado_FGV_Nicolas_Behar_Cambraia.pdf.txtExtracted texttext/plain100770https://repositorio.fgv.br/bitstreams/32794da4-e56f-4ba3-b8dd-81c7835f685d/download05d80ee1e83c9bc4dd8d43daef6e6e2bMD53falseAnonymousREADTHUMBNAILDissertação_Mestrado_FGV_Nicolas_Behar_Cambraia.pdf.jpgDissertação_Mestrado_FGV_Nicolas_Behar_Cambraia.pdf.jpgGenerated Thumbnailimage/jpeg2650https://repositorio.fgv.br/bitstreams/30b7fc35-b697-4bc6-b54c-e7d2fd1ee050/downloadbf79fc069e6c5d4d95afd0f6ad85d401MD54falseAnonymousREAD10438/372442025-07-29 15:22:25.667open.accessoai:repositorio.fgv.br:10438/37244https://repositorio.fgv.brRepositório InstitucionalPRIhttp://bibliotecadigital.fgv.br/dspace-oai/requestbiblioteca.digital@fgv.bropendoar:39742025-07-29T15:22:25Repositório Institucional do FGV (FGV Repositório Digital) - Fundação Getulio Vargas (FGV)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
dc.title.eng.fl_str_mv News sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock markets
title News sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock markets
spellingShingle News sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock markets
Cambraia, Nicolas Behar
HAR-X
Volatility forecasting
News sentiment
Previsão de volatilidade
Sentimento de notícias
Economia
Mercado de capitais - Brasil
Mercado de ações - Previsão
Volatilidade (Finanças)
Processo decisório
title_short News sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock markets
title_full News sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock markets
title_fullStr News sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock markets
title_full_unstemmed News sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock markets
title_sort News sentiment and volatility dynamics: a HAR-X analysis of Brazilian stock markets
author Cambraia, Nicolas Behar
author_facet Cambraia, Nicolas Behar
author_role author
dc.contributor.unidadefgv.por.fl_str_mv Escolas::EESP
dc.contributor.member.none.fl_str_mv Andrade, Douglas Coimbra de
dc.contributor.author.fl_str_mv Cambraia, Nicolas Behar
dc.contributor.advisor1.fl_str_mv Pinto, Afonso de Campos
Botelho, Marcos Antonio
contributor_str_mv Pinto, Afonso de Campos
Botelho, Marcos Antonio
dc.subject.eng.fl_str_mv HAR-X
Volatility forecasting
News sentiment
topic HAR-X
Volatility forecasting
News sentiment
Previsão de volatilidade
Sentimento de notícias
Economia
Mercado de capitais - Brasil
Mercado de ações - Previsão
Volatilidade (Finanças)
Processo decisório
dc.subject.por.fl_str_mv Previsão de volatilidade
Sentimento de notícias
dc.subject.area.por.fl_str_mv Economia
dc.subject.bibliodata.por.fl_str_mv Mercado de capitais - Brasil
Mercado de ações - Previsão
Volatilidade (Finanças)
Processo decisório
description The objective of this work is to apply to stocks traded on the Brazilian stock exchange the Heterogeneous Autoregressive model integrated with the use of news sentiment scores as an exogenous variable in order to estimate the effect of news on stock volatility. The volatility of the assets is obtained from the opening, minimum, maximum, and closing prices over one-hour intervals, using the Garman-Klass estimator. The sentiment scores are derived from a zero-shot classifier and proposed hypotheses, using as input approximately 2500 financial news from May 2023 to March 2025, with the intention of estimating the impact of news on stocks. Machine learning metrics such as R 2 and Mean Squared Error were compared between the models with and without exogenous variables for each evaluated asset. The inclusion of sentiment scores led to an overall improvement in performance. Furthermore, the analysis revealed a consistent lag structure across all evaluated stocks, characterized by similar linear coefficients associated with the lagged news sentiment variables. This pattern held irrespective of the industry in which each company operates, suggesting a generalized temporal response to news sentiment in volatility modeling.
publishDate 2025
dc.date.accessioned.fl_str_mv 2025-07-21T20:06:25Z
dc.date.available.fl_str_mv 2025-07-21T20:06:25Z
dc.date.issued.fl_str_mv 2025-06-24
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv https://hdl.handle.net/10438/37244
url https://hdl.handle.net/10438/37244
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.source.none.fl_str_mv reponame:Repositório Institucional do FGV (FGV Repositório Digital)
instname:Fundação Getulio Vargas (FGV)
instacron:FGV
instname_str Fundação Getulio Vargas (FGV)
instacron_str FGV
institution FGV
reponame_str Repositório Institucional do FGV (FGV Repositório Digital)
collection Repositório Institucional do FGV (FGV Repositório Digital)
bitstream.url.fl_str_mv https://repositorio.fgv.br/bitstreams/7b46cf49-1723-44b1-9bb5-0bf3a2b6cdf0/download
https://repositorio.fgv.br/bitstreams/84d2ac30-17e1-4fab-9333-ef37950678e6/download
https://repositorio.fgv.br/bitstreams/32794da4-e56f-4ba3-b8dd-81c7835f685d/download
https://repositorio.fgv.br/bitstreams/30b7fc35-b697-4bc6-b54c-e7d2fd1ee050/download
bitstream.checksum.fl_str_mv 6d61b7437ac819b9e744a1a3f174f875
2a4b67231f701c416a809246e7a10077
05d80ee1e83c9bc4dd8d43daef6e6e2b
bf79fc069e6c5d4d95afd0f6ad85d401
bitstream.checksumAlgorithm.fl_str_mv MD5
MD5
MD5
MD5
repository.name.fl_str_mv Repositório Institucional do FGV (FGV Repositório Digital) - Fundação Getulio Vargas (FGV)
repository.mail.fl_str_mv biblioteca.digital@fgv.br
_version_ 1866369147503378432

Registros relacionados