Earnings prediction using machine learning methods and analyst comparison
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | http://hdl.handle.net/10400.14/38654 |
Resumo: | In the course of this dissertation we propose an experimental study on how technical, macroeconomic, and financial variables, alongside analysts’ forecasts, can be used to optimize the prediction for the subsequent quarter’s earnings results using machine learning, comparing the performance of the models to analysts’ forecasts. The dissertation includes three steps. In step one, an event study is conducted to test abnormal returns in firms’ stock prices in the day following earnings announcement, grouped by earnings per share (EPS) growth in classes of size 3, 6 and 9, computed for each quarter. In step two, several machine learning models are built to maximize the accuracy of EPS predictions. In the last step, investment strategies are constructed to take advantage of investors’ expectations, which are closely correlated with analysts’ predictions. In the backdrop of an exhaustive analysis on quarterly earnings predictions using machine learning methods, conclusions are drawn related to the superiority of the CatBoost classifier. All machine learning models tested underperform analyst predictions, which could be explained by the time and privileged information at analysts’ disposal, as well as their selection of firms to cover. Regardless, machine learning models can be used as a confirmation for analyst predictions, and statistically significant investment strategies are pursued with those fundamentals. Importantly, high confidence predictions by machine learning models are significantly more accurate than the average accuracy of forecasts. |
id |
RCAP_3d394d3b262ff7826afe89daf8cdb1bc |
---|---|
oai_identifier_str |
oai:repositorio.ucp.pt:10400.14/38654 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Earnings prediction using machine learning methods and analyst comparisonEarnings announcementsAnalyst errorsEvent studyMachine learningTechnical analysisAnúncio de resultadosErros dos analistasEstudos de eventosAnálise técnicaDomínio/Área Científica::Ciências Médicas::Outras Ciências MédicasIn the course of this dissertation we propose an experimental study on how technical, macroeconomic, and financial variables, alongside analysts’ forecasts, can be used to optimize the prediction for the subsequent quarter’s earnings results using machine learning, comparing the performance of the models to analysts’ forecasts. The dissertation includes three steps. In step one, an event study is conducted to test abnormal returns in firms’ stock prices in the day following earnings announcement, grouped by earnings per share (EPS) growth in classes of size 3, 6 and 9, computed for each quarter. In step two, several machine learning models are built to maximize the accuracy of EPS predictions. In the last step, investment strategies are constructed to take advantage of investors’ expectations, which are closely correlated with analysts’ predictions. In the backdrop of an exhaustive analysis on quarterly earnings predictions using machine learning methods, conclusions are drawn related to the superiority of the CatBoost classifier. All machine learning models tested underperform analyst predictions, which could be explained by the time and privileged information at analysts’ disposal, as well as their selection of firms to cover. Regardless, machine learning models can be used as a confirmation for analyst predictions, and statistically significant investment strategies are pursued with those fundamentals. Importantly, high confidence predictions by machine learning models are significantly more accurate than the average accuracy of forecasts.No decorrer desta dissertação, realiza-se um estudo experimental sobre a forma como análises técnicas, macroeconómicas, fundamentais e as previsões dos analistas podem ser utilizadas em conjunto para otimizar a previsão dos resultados de lucros do próximo trimestre de empresas A dissertação inclui três etapas. Na primeira etapa, é efetuado um estudo de evento para testar os retornos anormais nas ações no dia seguinte aos anúncios de lucros, sendo estes agrupados pelo crescimento do lucro por ação nas classes de 3, 6 e 9, calculado para cada trimestre. Na etapa dois, vários modelos de machine learning (ML) são concebidos para maximizar a precisão das previsões de crescimento de lucros de empresas. Na última etapa, estratégias de investimento são construídas para tirar proveito das expectativas do investidor, que estão relacionadas com as previsões dos analistas. Uma vez que um dos projetos de pesquisa mais exaustivos sobre previsões de lucros para o próximo trimestre, conclusões podem ser retiradas relacionadas com a superioridade do modelo CatBoost nas previsões de lucros. Todos os modelos de testados apresentam desempenho inferior às previsões dos analistas, o que pode ser explicado pelo tempo e pelas informações privilegiadas a que os analistas têm acesso, bem como pela escolha da empresa sob a qual as suas previsões incidem. Os modelos de podem ser utilizados como uma confirmação para as previsões dos analistas criando estratégias de investimento estatisticamente significativas. Além disso, as previsões com alta confiança por modelos de são mais precisas do que a precisão média das previsões dos analistas.Trung, Tran HieuVeritati - Repositório Institucional da Universidade Católica PortuguesaMartins, Alexandre Inês2022-09-01T09:28:01Z2022-01-242022-012022-01-24T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10400.14/38654TID:202964558enginfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-07-12T17:44:09Zoai:repositorio.ucp.pt:10400.14/38654Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T18:31:34.158506Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Earnings prediction using machine learning methods and analyst comparison |
title |
Earnings prediction using machine learning methods and analyst comparison |
spellingShingle |
Earnings prediction using machine learning methods and analyst comparison Martins, Alexandre Inês Earnings announcements Analyst errors Event study Machine learning Technical analysis Anúncio de resultados Erros dos analistas Estudos de eventos Análise técnica Domínio/Área Científica::Ciências Médicas::Outras Ciências Médicas |
title_short |
Earnings prediction using machine learning methods and analyst comparison |
title_full |
Earnings prediction using machine learning methods and analyst comparison |
title_fullStr |
Earnings prediction using machine learning methods and analyst comparison |
title_full_unstemmed |
Earnings prediction using machine learning methods and analyst comparison |
title_sort |
Earnings prediction using machine learning methods and analyst comparison |
author |
Martins, Alexandre Inês |
author_facet |
Martins, Alexandre Inês |
author_role |
author |
dc.contributor.none.fl_str_mv |
Trung, Tran Hieu Veritati - Repositório Institucional da Universidade Católica Portuguesa |
dc.contributor.author.fl_str_mv |
Martins, Alexandre Inês |
dc.subject.por.fl_str_mv |
Earnings announcements Analyst errors Event study Machine learning Technical analysis Anúncio de resultados Erros dos analistas Estudos de eventos Análise técnica Domínio/Área Científica::Ciências Médicas::Outras Ciências Médicas |
topic |
Earnings announcements Analyst errors Event study Machine learning Technical analysis Anúncio de resultados Erros dos analistas Estudos de eventos Análise técnica Domínio/Área Científica::Ciências Médicas::Outras Ciências Médicas |
description |
In the course of this dissertation we propose an experimental study on how technical, macroeconomic, and financial variables, alongside analysts’ forecasts, can be used to optimize the prediction for the subsequent quarter’s earnings results using machine learning, comparing the performance of the models to analysts’ forecasts. The dissertation includes three steps. In step one, an event study is conducted to test abnormal returns in firms’ stock prices in the day following earnings announcement, grouped by earnings per share (EPS) growth in classes of size 3, 6 and 9, computed for each quarter. In step two, several machine learning models are built to maximize the accuracy of EPS predictions. In the last step, investment strategies are constructed to take advantage of investors’ expectations, which are closely correlated with analysts’ predictions. In the backdrop of an exhaustive analysis on quarterly earnings predictions using machine learning methods, conclusions are drawn related to the superiority of the CatBoost classifier. All machine learning models tested underperform analyst predictions, which could be explained by the time and privileged information at analysts’ disposal, as well as their selection of firms to cover. Regardless, machine learning models can be used as a confirmation for analyst predictions, and statistically significant investment strategies are pursued with those fundamentals. Importantly, high confidence predictions by machine learning models are significantly more accurate than the average accuracy of forecasts. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-09-01T09:28:01Z 2022-01-24 2022-01 2022-01-24T00:00:00Z |
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 |
http://hdl.handle.net/10400.14/38654 TID:202964558 |
url |
http://hdl.handle.net/10400.14/38654 |
identifier_str_mv |
TID:202964558 |
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.format.none.fl_str_mv |
application/pdf |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799132039421624320 |