Edson Vinicius Pontes Bastos
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Edson Vinicius Pontes Bastos has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Luciana Holtz
Universidade Federal de Juiz de Fora (UFJF)
2 shared publications - Marcelo Alvaro da Silva Macedo
Universidade Federal do Rio de Janeiro (PPGCC/UFRJ)
2 shared publications - Odilanei Morais dos Santos
Universidade Federal do Rio de Janeiro (UFRJ)
2 shared publications Jorge Junio Moreira Antunes
Univ Fed Rio De Janeiro UFRJ, Inst Pos Grad Adm Coppead, BR-21941918 Rio De Janeiro, Brazil
1 shared publicationLarissa Pereira DE Farias
PEP/COPPE/UFRJ
1 shared publication- Lino Guimarães Marujo
Universidade Federal do Rio de Janeiro
1 shared publication
Latest Publications
- Open Access
- 12 September 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (18) , 3318
Statistical Accuracy, Economic Value and Model Instability in ETF Return Forecasting: A Comparison Across Developed and Emerging Markets
Whether machine-learning models extract predictive signal from ETF returns across markets at different efficiency levels and whether statistical accuracy translates into trading value, remain contested. We examine this for iShares MSCI Brazil (EWZ) and iShares Core S&P 500 (IVV) from January 2010 to July 2026, training through December 2022 and testing thereafter. Random Forest, XGBoost with random search, XGBoost with Bayesian optimization, LSTM, GRU and an LSTM + XGBoost ensemble, are compared against historical mean, random walk, and AR(1) benchmarks at one-day (h = 1), five-day (h = 5) and monthly (h = 21) horizons using ten technical indicators. Every model is also evaluated against the classifier that predicts the majority class, and risk-adjusted performance is reported with bootstrap intervals. No model exceeds that trivial classifier in any combination examined. The two markets fail by distinct mechanisms: collapse onto the majority class in the developed market, and dispersed but unprofitable signals in the emerging one. Under Diebold–Mariano tests with autocorrelation-consistent variance and false-discovery control, no model is superior to the historical mean. No strategy outperforms Buy-and-Hold, and in the emerging market, no Sharpe ratio is distinguishable from zero. Where directional significance does appear, at the monthly horizon in the emerging market, it delivers no economic value. Together, these results argue for evaluating financial forecasting models simultaneously on regression metrics, economic performance, and regime stability rather than on any single criterion.
- 12 September 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (18) , 3318
Open AccessStatistical Accuracy, Economic Value and Model Instability in ETF Return Forecasting: A Comparison Across Developed and Emerging Markets
Whether machine-learning models extract predictive signal from ETF returns across markets at different efficiency levels and whether statistical accuracy translates into trading value, remain contested. We examine this for iShares MSCI Brazil (EWZ) and iShares Core S&P 500 (IVV) from January 2010 to July 2026, training through December 2022 and testing thereafter. Random Forest, XGBoost with random search, XGBoost with Bayesian optimization, LSTM, GRU and an LSTM + XGBoost ensemble, are compared against historical mean, random walk, and AR(1) benchmarks at one-day (h = 1), five-day (h = 5) and monthly (h = 21) horizons using ten technical indicators. Every model is also evaluated against the classifier that predicts the majority class, and risk-adjusted performance is reported with bootstrap intervals. No model exceeds that trivial classifier in any combination examined. The two markets fail by distinct mechanisms: collapse onto the majority class in the developed market, and dispersed but unprofitable signals in the emerging one. Under Diebold–Mariano tests with autocorrelation-consistent variance and false-discovery control, no model is superior to the historical mean. No strategy outperforms Buy-and-Hold, and in the emerging market, no Sharpe ratio is distinguishable from zero. Where directional significance does appear, at the monthly horizon in the emerging market, it delivers no economic value. Together, these results argue for evaluating financial forecasting models simultaneously on regression metrics, economic performance, and regime stability rather than on any single criterion.
- Open Access
- 1 January 2023
- Published by Tech Science Press in Intelligent Automation and Soft Computing
- Vol. 36 (3) , 3175-3190
An Endogenous Feedback and Entropy Analysis in Machine Learning Model for Stock’s Return Forecast
Stock markets exhibit Brownian movement with random, non-linear, uncertain, evolutionary, non-parametric, nebulous, chaotic characteristics and dynamism with a high degree of complexity. Developing an algorithm to predict returns for decision-making is a challenging goal. In addition, the choice of variables that will serve as input to the model represents a non-triviality, since it is possible to observe endogeneity problems between the predictor and the predicted variables. Thus, the goal is to analyze the endogenous origin of the stock return prediction model based on technical indicators. For this, we structure a feed-forward neural network. We evaluate the endogenous feedback between the predicted returns and technical analysis indicators based on the generated residues. It is possible to predict the return. The high accuracy of the model indicates that, during the test period, there is a hit rate close to 76%. Regarding endogeneity, the term of interest and the return are the variables that influence the largest number of indicators. The results will help investors build investment strategies based on this expert system applied to forecasting.
- 1 January 2023
- Published by Tech Science Press in Intelligent Automation and Soft Computing
- Vol. 36 (3) , 3175-3190
Open AccessAn Endogenous Feedback and Entropy Analysis in Machine Learning Model for Stock’s Return Forecast
Stock markets exhibit Brownian movement with random, non-linear, uncertain, evolutionary, non-parametric, nebulous, chaotic characteristics and dynamism with a high degree of complexity. Developing an algorithm to predict returns for decision-making is a challenging goal. In addition, the choice of variables that will serve as input to the model represents a non-triviality, since it is possible to observe endogeneity problems between the predictor and the predicted variables. Thus, the goal is to analyze the endogenous origin of the stock return prediction model based on technical indicators. For this, we structure a feed-forward neural network. We evaluate the endogenous feedback between the predicted returns and technical analysis indicators based on the generated residues. It is possible to predict the return. The high accuracy of the model indicates that, during the test period, there is a hit rate close to 76%. Regarding endogeneity, the term of interest and the return are the variables that influence the largest number of indicators. The results will help investors build investment strategies based on this expert system applied to forecasting.
- Open Access
- 31 January 2022
- Published by Revista Produção e Desenvolvimento in Revista Produção e Desenvolvimento
- Vol. 8 (1) , e612
Modelo contábil fleuriet para a tomada de decisão nos subsetores agropecuários e de alimentos processados brasileiros
Purpose: this article aims to apply a model to assess the financial and accounting situation of companies in the selected Agricultural and Processed Food and Livestock economic subsectors, using accounting analysis tools and statistical support identify correlations between the fleuriet dynamic model indicators with traditional indicators and stock market indicators. Methodology/Approach: This research is framed as a documental study with B3 portal information at the national level of companies operating in Brazil. Findings: The dynamic model indicators are correlated with both traditional indicators and stock market indicators, but with the exception of a few isolated cases, the ROI and ROE indicators did not have such a high correlation rate with the others. There was a high correlation between the indicators CDG, NDG and T, with positive and negative variations depending on the studied typology. Research Limitation/implication: can register not full disponible of the company's market values for the survey recent years so that the statistical analysis could be carried out using these data. Originality/Value of paper: Fleuriet's dynamic model appears can be used in order to identify the real financial situation of organizations and thus generate more confidence in managers for decision-making.
- 31 January 2022
- Published by Revista Produção e Desenvolvimento in Revista Produção e Desenvolvimento
- Vol. 8 (1) , e612
Open AccessModelo contábil fleuriet para a tomada de decisão nos subsetores agropecuários e de alimentos processados brasileiros
Purpose: this article aims to apply a model to assess the financial and accounting situation of companies in the selected Agricultural and Processed Food and Livestock economic subsectors, using accounting analysis tools and statistical support identify correlations between the fleuriet dynamic model indicators with traditional indicators and stock market indicators. Methodology/Approach: This research is framed as a documental study with B3 portal information at the national level of companies operating in Brazil. Findings: The dynamic model indicators are correlated with both traditional indicators and stock market indicators, but with the exception of a few isolated cases, the ROI and ROE indicators did not have such a high correlation rate with the others. There was a high correlation between the indicators CDG, NDG and T, with positive and negative variations depending on the studied typology. Research Limitation/implication: can register not full disponible of the company's market values for the survey recent years so that the statistical analysis could be carried out using these data. Originality/Value of paper: Fleuriet's dynamic model appears can be used in order to identify the real financial situation of organizations and thus generate more confidence in managers for decision-making.
- Open Access
- 3 January 2022
- Published by Editora Unijui in Revista Contexto & Educação
- Vol. 37 (116) , 226-238
ANÁLISE DA RELAÇÃO ENTRE A REMUNERAÇÃO DO TRABALHO E O NÍVEL DE INSTRUÇÃO: UMA ABORDAGEM ESTATÍSTICA.
Há uma concordância geral indicativa de que o nível de escolaridade é o principal fator que determina as discrepâncias de renda salarial entre indivíduos. Entretanto, não há um consenso a respeito do poder explicativo dos diferentes níveis de escolaridade e seu impacto no salário de um indivíduo. Desse modo, objetiva-se verificar se o poder explicativo da variável grau de instrução sobre a remuneração sofre acréscimo na proporção em que se aumentam os anos de estudo, indicando, portanto, alguns níveis onde outras variáveis influenciariam mais a remuneração do que o nível de escolaridade. Os resultados sugeriram que para os grupos de menor escolaridade o poder explicativo dos anos de estudo sobre a remuneração é estatisticamente significativo, porém é inferior ao poder explicativo sobre o grupo com maior nível de escolaridade, indicando que para os grupos de menor escolaridade existe outros fatores que impactam em maior grau a remuneração do indivíduo.
- 3 January 2022
- Published by Editora Unijui in Revista Contexto & Educação
- Vol. 37 (116) , 226-238
Open AccessANÁLISE DA RELAÇÃO ENTRE A REMUNERAÇÃO DO TRABALHO E O NÍVEL DE INSTRUÇÃO: UMA ABORDAGEM ESTATÍSTICA.
Há uma concordância geral indicativa de que o nível de escolaridade é o principal fator que determina as discrepâncias de renda salarial entre indivíduos. Entretanto, não há um consenso a respeito do poder explicativo dos diferentes níveis de escolaridade e seu impacto no salário de um indivíduo. Desse modo, objetiva-se verificar se o poder explicativo da variável grau de instrução sobre a remuneração sofre acréscimo na proporção em que se aumentam os anos de estudo, indicando, portanto, alguns níveis onde outras variáveis influenciariam mais a remuneração do que o nível de escolaridade. Os resultados sugeriram que para os grupos de menor escolaridade o poder explicativo dos anos de estudo sobre a remuneração é estatisticamente significativo, porém é inferior ao poder explicativo sobre o grupo com maior nível de escolaridade, indicando que para os grupos de menor escolaridade existe outros fatores que impactam em maior grau a remuneração do indivíduo.
- Open Access
- 30 October 2021
- Published by Associacao Brasileira de Engenharia de Producao - ABEPRO
COLABORAÇÃO TECNOLÓGICA NO ENFRENTAMENTO DA PANDEMIA DE COVID-19: VENTILADORES PULMONARES E O CASO VEXCO
- 30 October 2021
- Published by Associacao Brasileira de Engenharia de Producao - ABEPRO
Open AccessCOLABORAÇÃO TECNOLÓGICA NO ENFRENTAMENTO DA PANDEMIA DE COVID-19: VENTILADORES PULMONARES E O CASO VEXCO
- Open Access
- 2 July 2021
- Published by Universidade Federal do Rio Grande do Norte - UFRN in Revista Ambiente ContÁbil - Universidade Federal Do Rio Grande Do Norte
- Vol. 13 (2) , 80-99
Mensuração a valor justo e seus impactos nos honorários de auditoria: evidências no mercado brasileiro
Objetivo: O objetivo é verificar o impacto do uso da mensuração a valor justo sobre os honorários de auditoria, diferenciando-se inclusive o período pré e pós adoção do CPC 46. Metodologia: A pesquisa apresentada é quantitativa, sendo que para o teste das hipóteses levantadas foi utilizada a técnica de regressão múltipla, com dados disponíveis das empresas listadas na B3 referentes ao período entre 2010 e 2016. Resultados: As evidências indicam que a complexidade e subjetividade do valor justo é reconhecida pelas empresas de auditoria, ou seja, as firmas de auditoria consideram que a mensuração a valor justo implica mais esforços e que os riscos de auditoria associados se elevam, levando as firmas à cobrança de um prêmio de risco pela prestação do serviço. Entretanto, não foi possível confirmar que os honorários dos auditores aumentaram após a adoção da IFRS13 (CPC 46). Contribuições do estudo: Teóricas/metodológicas - o estudo contribui para a compreensão sobre os impactos da adoção de normas internacionais de contabilidade, neste caso específico, sobre honorários de auditoria. Sociais/para gestão - dada a evidência de que há maior custo de auditoria associado à maior complexidade das informações em ambiente de valor justo, as empresas podem desenvolver mecanismos de minimização de incerteza associada às informações a serem auditadas.
- 2 July 2021
- Published by Universidade Federal do Rio Grande do Norte - UFRN in Revista Ambiente ContÁbil - Universidade Federal Do Rio Grande Do Norte
- Vol. 13 (2) , 80-99
Open AccessMensuração a valor justo e seus impactos nos honorários de auditoria: evidências no mercado brasileiro
Objetivo: O objetivo é verificar o impacto do uso da mensuração a valor justo sobre os honorários de auditoria, diferenciando-se inclusive o período pré e pós adoção do CPC 46. Metodologia: A pesquisa apresentada é quantitativa, sendo que para o teste das hipóteses levantadas foi utilizada a técnica de regressão múltipla, com dados disponíveis das empresas listadas na B3 referentes ao período entre 2010 e 2016. Resultados: As evidências indicam que a complexidade e subjetividade do valor justo é reconhecida pelas empresas de auditoria, ou seja, as firmas de auditoria consideram que a mensuração a valor justo implica mais esforços e que os riscos de auditoria associados se elevam, levando as firmas à cobrança de um prêmio de risco pela prestação do serviço. Entretanto, não foi possível confirmar que os honorários dos auditores aumentaram após a adoção da IFRS13 (CPC 46). Contribuições do estudo: Teóricas/metodológicas - o estudo contribui para a compreensão sobre os impactos da adoção de normas internacionais de contabilidade, neste caso específico, sobre honorários de auditoria. Sociais/para gestão - dada a evidência de que há maior custo de auditoria associado à maior complexidade das informações em ambiente de valor justo, as empresas podem desenvolver mecanismos de minimização de incerteza associada às informações a serem auditadas.
- Open Access
- 2 July 2021
- Published by Universidade Federal do Rio Grande do Norte - UFRN in Revista Ambiente ContÁbil - Universidade Federal Do Rio Grande Do Norte
- Vol. 13 (2) , 80-99
Fair value measurement and its impact on audit fees: evidence in the Brazilian market
Objective: The purpose is to verify the impact of using the measurement at fair value on the audit fees, differentiating even the period before and after the adoption of IFRS 13 (CPC 46). Methodology: The research is quantitative, for testing the hypothesis raised, the multiple regression technique was used, with data available from companies listed in B3 for the period between 2010 and 2016. Results: The evidence indicates that the complexity and subjectivity of fair value is recognized by the audit firms, that is, audit firms recognize that fair value measurement implies more effort and that the associated audit risk rises, leading firms to charge of a risk premium for the provision of the service. However, it was not possible to confirm that auditors' fees increased after the adoption of IFRS13 (CPC 46). Contributions of study: Theoretical/methodological - The study contributes to understanding the impacts of adopting international accounting standards, in this specific case on audit fees. Social/management - Given the evidence that there is a higher audit cost associated with the greater complexity of information in a fair value environment, companies can develop mechanisms to minimize the uncertainty of the information to be audited.
- 2 July 2021
- Published by Universidade Federal do Rio Grande do Norte - UFRN in Revista Ambiente ContÁbil - Universidade Federal Do Rio Grande Do Norte
- Vol. 13 (2) , 80-99
Open AccessFair value measurement and its impact on audit fees: evidence in the Brazilian market
Objective: The purpose is to verify the impact of using the measurement at fair value on the audit fees, differentiating even the period before and after the adoption of IFRS 13 (CPC 46). Methodology: The research is quantitative, for testing the hypothesis raised, the multiple regression technique was used, with data available from companies listed in B3 for the period between 2010 and 2016. Results: The evidence indicates that the complexity and subjectivity of fair value is recognized by the audit firms, that is, audit firms recognize that fair value measurement implies more effort and that the associated audit risk rises, leading firms to charge of a risk premium for the provision of the service. However, it was not possible to confirm that auditors' fees increased after the adoption of IFRS13 (CPC 46). Contributions of study: Theoretical/methodological - The study contributes to understanding the impacts of adopting international accounting standards, in this specific case on audit fees. Social/management - Given the evidence that there is a higher audit cost associated with the greater complexity of information in a fair value environment, companies can develop mechanisms to minimize the uncertainty of the information to be audited.
- Open Access
- 31 December 2020
- Published by Revista Produção e Desenvolvimento in Revista Produção e Desenvolvimento
- Vol. 6
Use of fleuriet model for forecasting and evaluating municipal electoral results in a brazilian state
Objective: Verify the relation between the variables in the Fleuriet model and the electoral result given by the exchange or maintenance of the party of the prefectures under analysis. Methodology: Through logistic regression it is possible to explain or predict the probability of the occurrence of the event under analysis, such event discussed here refers to the exchange of the party, which is a dummy variable that assumes a value of 1 when there is an exchange and zero otherwise, the method of estimation used to obtain the coefficients was based on maximum likelihood. Results: The relationship between the variables of the Fleuriet model and the electoral result given by the exchange or maintenance of the party in the prefectures of the group under analysis. Limitation or implication of the research: It is noteworthy that such research does not allow generalizations, being a study with reduced scope considering only the city halls of the State of Rio de Janeiro that have financial information disclosed. Originality: showing that city halls with better financial situation tend not to change the party, which was observed in the group under analysis considering the election event in 2020.
- 31 December 2020
- Published by Revista Produção e Desenvolvimento in Revista Produção e Desenvolvimento
- Vol. 6
Open AccessUse of fleuriet model for forecasting and evaluating municipal electoral results in a brazilian state
Objective: Verify the relation between the variables in the Fleuriet model and the electoral result given by the exchange or maintenance of the party of the prefectures under analysis. Methodology: Through logistic regression it is possible to explain or predict the probability of the occurrence of the event under analysis, such event discussed here refers to the exchange of the party, which is a dummy variable that assumes a value of 1 when there is an exchange and zero otherwise, the method of estimation used to obtain the coefficients was based on maximum likelihood. Results: The relationship between the variables of the Fleuriet model and the electoral result given by the exchange or maintenance of the party in the prefectures of the group under analysis. Limitation or implication of the research: It is noteworthy that such research does not allow generalizations, being a study with reduced scope considering only the city halls of the State of Rio de Janeiro that have financial information disclosed. Originality: showing that city halls with better financial situation tend not to change the party, which was observed in the group under analysis considering the election event in 2020.
- 4 November 2020
- Published by Fucape Business School in Brazilian Business Review
- Vol. 17 (6) , 621-639
Fundamentalist Signals in Volatility Scenarios: Evidence in the Brazilian Stock Market
This article investigates whether the usefulness of fundamentalist signals to predict returns are altered in context of high volatility and also considering the sensitivity of assets to the IVol-BR volatility index. In times of high volatility, investors could make their decisions based on risk aversion and not only on the fundamentals signals of companies. In addition, it is possible to see how different delays in fundamentalist signals are related to future returns. The methodological choice is for estimators in panel data for the analysis of non-financial companies that have shares traded on B3 - Brasil, Bolsa, Balcão – in the period from 2011.3Q to 2018.2Q. The results show evidence of changes in the explanatory capacity of fundamentalist signals in different volatility scenarios, and for different sensitivities to IVol-BR. This finding may impact the decision-making of managers and investors as it enables the design of investment strategies based on fundamentalist signals adhering to different risk scenarios.
- 4 November 2020
- Published by Fucape Business School in Brazilian Business Review
- Vol. 17 (6) , 621-639
Fundamentalist Signals in Volatility Scenarios: Evidence in the Brazilian Stock Market
This article investigates whether the usefulness of fundamentalist signals to predict returns are altered in context of high volatility and also considering the sensitivity of assets to the IVol-BR volatility index. In times of high volatility, investors could make their decisions based on risk aversion and not only on the fundamentals signals of companies. In addition, it is possible to see how different delays in fundamentalist signals are related to future returns. The methodological choice is for estimators in panel data for the analysis of non-financial companies that have shares traded on B3 - Brasil, Bolsa, Balcão – in the period from 2011.3Q to 2018.2Q. The results show evidence of changes in the explanatory capacity of fundamentalist signals in different volatility scenarios, and for different sensitivities to IVol-BR. This finding may impact the decision-making of managers and investors as it enables the design of investment strategies based on fundamentalist signals adhering to different risk scenarios.
- Open Access
- 11 November 2019
- Published by Fundacao Getulio Vargas in Revista de Gestão dos Países de Língua Portuguesa
- Vol. 18 (2) , 104-125
Relevância da Informação Contábil de Lucro e Fluxo de Caixa: um Estudo com Foco em Ações Ordinárias e Preferenciais
Objetiva-se verificar se o lucro líquido por ação (LPA) e o fluxo de caixa operacional por ação (FCO_A) apresentam capacidades distintas de explicar o comportamento dos preços das ações ordinárias e preferenciais. Formaram-se três grupos: o primeiro, com ambas as classes acionárias; o segundo, apenas com ações ordinárias; e o terceiro, apenas com preferenciais. Para cada grupo foram aplicados dois modelos de regressões, um usando o LPA e outro usando o FCO_A, tendo em ambos o patrimônio líquido, QTobin e o tamanho como variável de controle. Os modelos foram estimados por OLS e, posteriormente, por GMM como análise de robustez. Os resultados sugerem que o FCO é mais relevante para as ações ordinárias do que para as ações preferenciais. Já a variável LPA é significativa para ambas as classes acionárias, e seus modelos apresentaram maior capacidade explicativa para as ações preferenciais. Logo, têm-se evidências de que o lucro líquido (LL) é mais relevante para as ações preferenciais do que para as ações ordinárias.
- 11 November 2019
- Published by Fundacao Getulio Vargas in Revista de Gestão dos Países de Língua Portuguesa
- Vol. 18 (2) , 104-125
Open AccessRelevância da Informação Contábil de Lucro e Fluxo de Caixa: um Estudo com Foco em Ações Ordinárias e Preferenciais
Objetiva-se verificar se o lucro líquido por ação (LPA) e o fluxo de caixa operacional por ação (FCO_A) apresentam capacidades distintas de explicar o comportamento dos preços das ações ordinárias e preferenciais. Formaram-se três grupos: o primeiro, com ambas as classes acionárias; o segundo, apenas com ações ordinárias; e o terceiro, apenas com preferenciais. Para cada grupo foram aplicados dois modelos de regressões, um usando o LPA e outro usando o FCO_A, tendo em ambos o patrimônio líquido, QTobin e o tamanho como variável de controle. Os modelos foram estimados por OLS e, posteriormente, por GMM como análise de robustez. Os resultados sugerem que o FCO é mais relevante para as ações ordinárias do que para as ações preferenciais. Já a variável LPA é significativa para ambas as classes acionárias, e seus modelos apresentaram maior capacidade explicativa para as ações preferenciais. Logo, têm-se evidências de que o lucro líquido (LL) é mais relevante para as ações preferenciais do que para as ações ordinárias.