# Marcelo S. Perlin — Full Site Reference > Associate Professor of Finance, Federal University of Rio Grande do Sul (UFRGS) (Porto Alegre, Brazil). Associate Professor of Finance at the Federal University of Rio Grande do Sul (UFRGS), Porto Alegre (Brazil), with a passion for quantitative finance and data science. My work focuses on developing and applying computational tools to solve real-world financial problems. I am the author of several books and research papers on data analysis, financial markets, and programming in R. I also develop and maintain open-source software in R and Python. For more details on my research and teaching, please see the Publications and Code sections. - **Website:** https://www.msperlin.com/ - **Email:** marceloperlin@gmail.com - **ORCID:** https://orcid.org/0000-0002-9839-4268 - **Google Scholar:** 1693 citations, h-index 18, i10-index 25 (updated 2026-11-08) - **GitHub:** https://github.com/msperlin ## Research Themes ### Market microstructure and liquidity How prices form in Brazilian markets, how liquidity varies across assets and venues, and what high-frequency data reveal about trading and market quality. Packages: BatchGetSymbols ### Financial econometrics and volatility Time-series models for volatility, risk forecasting and the empirical behaviour of financial returns. Packages: BatchGetSymbols, GetTDData ### Science of science and publishing Bibliometrics, predatory publishing, academic inbreeding and the incentives that shape scientific production. Packages: GetLattesData ### Open data, reproducibility and software Building open-source tools that make Brazilian financial and economic data accessible, reproducible and easy to analyse in R and Python. Packages: GetDFPData2, GetFREData, GetTDData, cvm-mcp, rfbpy ### Artificial intelligence and LLMs in finance How large language models and machine learning are changing financial analysis, research and academic publishing. Packages: finbert-ai-detector ## Brazilian Financial Data MCP Servers Open-source Model Context Protocol (MCP) servers that give AI assistants direct, structured access to official Brazilian financial, corporate, and macroeconomic open data — with local caching and zero API keys required. Hub URL: https://www.msperlin.com/mcp ### cvm-mcp (CVM / B3 Open Data) An MCP server for the CVM open-data repository (dados.cvm.gov.br), giving MCP clients one interface to Brazilian financial statements, corporate filings, fund data, insider trading and more. - Detail page: https://www.msperlin.com/code/python/cvm-mcp - Markdown: https://www.msperlin.com/code/python/cvm-mcp.md - Run via uvx: `uvx cvm-mcp` - Annual (DFP) and quarterly (ITR) financial statements of Brazilian public companies - Corporate reference filings (FRE), governance, and corporate events - Brazilian investment funds (fundos de investimento) data and registration - Insider trading reports and CVM public open-data archives (dados.cvm.gov.br) ### bcb-mcp (Central Bank of Brazil (BCB-SGS)) An MCP server for the time series of the Central Bank of Brazil (BCB-SGS), with a curated series dictionary, a local parquet cache and DuckDB SQL for LLM agents. It is the Python counterpart of the R package GetBCBData. - Detail page: https://www.msperlin.com/code/python/bcb-mcp - Markdown: https://www.msperlin.com/code/python/bcb-mcp.md - Run via uvx: `uvx bcb-mcp` - Curated dictionary and search across BCB-SGS macroeconomic and financial series (Selic, IPCA, CDI, PTAX, GDP, credit) - Time-series retrieval and summary statistics directly from the official Central Bank API - Local Parquet caching with DuckDB SQL querying for fast multi-series analysis ### tesourodireto-mcp (Tesouro Nacional / ANBIMA) An MCP server for Brazilian government bonds (Tesouro Direto): historical prices and yields, current rates, per-maturity statistics and the ANBIMA yield curve. It is the Python counterpart of the R package GetTDData. - Detail page: https://www.msperlin.com/code/python/tesourodireto-mcp - Markdown: https://www.msperlin.com/code/python/tesourodireto-mcp.md - Run via uvx: `uvx tesourodireto-mcp` - Historical daily prices and yields for Brazilian sovereign bonds (Tesouro Direto) - Current Tesouro Direto purchase and redemption rates across available maturities - Per-maturity yield statistics and the ANBIMA sovereign yield curve ## Open-Source Software (R & Python) ### BatchGetSymbols (R) Downloads and Organizes Financial Data for Multiple Tickers - Detail page: https://www.msperlin.com/code/r/batchgetsymbols - Markdown: https://www.msperlin.com/code/r/batchgetsymbols.md - Install: `install.packages('BatchGetSymbols')` - Downloads: 714 last month / 193,127 total - Repository: https://github.com/msperlin/BatchGetSymbols ### GetBCBData (R) Imports Datasets from BCB (Central Bank of Brazil) using Its Official API - Detail page: https://www.msperlin.com/code/r/getbcbdata - Markdown: https://www.msperlin.com/code/r/getbcbdata.md - Install: `install.packages('GetBCBData')` - Downloads: 831 last month / 63,821 total - Repository: https://github.com/msperlin/GetBCBData ### GetDFPData2 (R) Reading Annual and Quarterly Financial Reports from B3 - Detail page: https://www.msperlin.com/code/r/getdfpdata2 - Markdown: https://www.msperlin.com/code/r/getdfpdata2.md - Install: `install.packages('GetDFPData2')` - Downloads: 596 last month / 42,457 total - Repository: https://github.com/msperlin/GetDFPData2 ### GetFREData (R) Reading FRE Corporate Data of Public Traded Companies from B3 - Detail page: https://www.msperlin.com/code/r/getfredata - Markdown: https://www.msperlin.com/code/r/getfredata.md - Install: `install.packages('GetFREData')` - Downloads: 415 last month / 17,907 total - Repository: https://github.com/msperlin/GetFREData ### GetLattesData (R) Reading Bibliometric Data from Lattes Platform - Detail page: https://www.msperlin.com/code/r/getlattesdata - Markdown: https://www.msperlin.com/code/r/getlattesdata.md - Install: `install.packages('GetLattesData')` - Downloads: 590 last month / 49,023 total - Repository: https://github.com/msperlin/GetLattesData ### GetQuandlData (R) Fast and Cached Import of Data from 'Quandl' Using the 'json API' - Detail page: https://www.msperlin.com/code/r/getquandldata - Markdown: https://www.msperlin.com/code/r/getquandldata.md - Install: `install.packages('GetQuandlData')` - Downloads: 383 last month / 26,991 total ### GetTDData (R) Get Data for Brazilian Bonds (Tesouro Direto) - Detail page: https://www.msperlin.com/code/r/gettddata - Markdown: https://www.msperlin.com/code/r/gettddata.md - Install: `install.packages('GetTDData')` - Downloads: 954 last month / 85,304 total - Repository: https://github.com/msperlin/GetTDData ### eodhdR2 (R) Official R API for Fetching Data from 'EODHD' - Detail page: https://www.msperlin.com/code/r/eodhdr2 - Markdown: https://www.msperlin.com/code/r/eodhdr2.md - Install: `install.packages('eodhdR2')` - Downloads: 436 last month / 7,818 total ### yfR (R) Downloads and Organizes Financial Data from Yahoo Finance - Detail page: https://www.msperlin.com/code/r/yfr - Markdown: https://www.msperlin.com/code/r/yfr.md - Install: `install.packages('yfR')` - Downloads: 1,087 last month / 59,257 total - Repository: https://github.com/ropensci/yfR ### bcb-mcp (Python) An MCP server for the time series of the Central Bank of Brazil (BCB-SGS), with a curated series dictionary, a local parquet cache and DuckDB SQL for LLM agents. It is the Python counterpart of the R package GetBCBData. - Detail page: https://www.msperlin.com/code/python/bcb-mcp - Markdown: https://www.msperlin.com/code/python/bcb-mcp.md - Install: `pip install bcb-mcp` - Downloads: 142 last month / 142 total - Repository: https://github.com/msperlin/bcb-mcp ### cvm-mcp (Python) An MCP server for the CVM open-data repository (dados.cvm.gov.br), giving MCP clients one interface to Brazilian financial statements, corporate filings, fund data, insider trading and more. - Detail page: https://www.msperlin.com/code/python/cvm-mcp - Markdown: https://www.msperlin.com/code/python/cvm-mcp.md - Install: `pip install cvm-mcp` - Downloads: 370 last month / 370 total - Repository: https://github.com/msperlin/cvm-mcp ### finbert-ai-detector (Python) An easy-to-use Python package for the msperlin/finbert-ai-detector model, designed to detect AI-generated text in financial documents. - Detail page: https://www.msperlin.com/code/python/finbert-ai-detector - Markdown: https://www.msperlin.com/code/python/finbert-ai-detector.md - Install: `pip install finbert-ai-detector` - Downloads: 42 last month / 2,327 total - Repository: https://github.com/msperlin/finbert-ai-detector ### rfbpy (Python) An easy-to-use Python package to download, extract and load the open CNPJ (Brazilian company registry) data published by the Receita Federal into SQLite or parquet files. - Detail page: https://www.msperlin.com/code/python/rfbpy - Markdown: https://www.msperlin.com/code/python/rfbpy.md - Install: `pip install rfbpy` - Downloads: 144 last month / 463 total - Repository: https://github.com/msperlin/rfbpy ### tesourodireto-mcp (Python) An MCP server for Brazilian government bonds (Tesouro Direto): historical prices and yields, current rates, per-maturity statistics and the ANBIMA yield curve. It is the Python counterpart of the R package GetTDData. - Detail page: https://www.msperlin.com/code/python/tesourodireto-mcp - Markdown: https://www.msperlin.com/code/python/tesourodireto-mcp.md - Install: `pip install tesourodireto-mcp` - Downloads: 139 last month / 139 total - Repository: https://github.com/msperlin/tesourodireto-mcp ## Books ### Poupando e Investindo em Renda Fixa: Uma Abordagem Baseada em Dados (2019) Uma população que não poupa e não sabe investir está fadada a continuar presa no ciclo vicioso da dependência financeira - mais gastos, mais trabalho e menos tempo. A falta de intimidade da população com a organização de um orçamento doméstico e com a aplicação do dinheiro de forma eficiente é certamente um dos problemas estruturais da economia brasileira. Este livro apresenta uma forma de investir com foco no longo prazo, resiliência e na qualidade de vida do investidor. Contrário a crença popular, investir para o longo prazo é fácil e pode ser feito por qualquer pessoa com vontade de aprender. O grande diferencial da obra é o uso frequente de dados financeiros reais para entender melhor como o mercado de renda fixa funciona, os efeitos dos tributos e custos operacionais, e quais são as melhores opções para o investidor pessoa física. - Detail page: https://www.msperlin.com/book/2019-pirf - Read online: https://www.msperlin.com/books/pirf/ - Amazon: https://www.amazon.com.br/dp/B07RR9K9PV ### Análise de Dados Financeiros e Econômicos com o R (Terceira Edição) (2021) Este livro introduz o leitor ao uso do R e RStudio como plataforma de análise de dados financeiros e econômicos. O livro apresenta toda a base de conhecimento necessária para utilizar o R, desde a sua instalação até a criação de códigos de pesquisa. O livro está organizado com exemplos práticos de uso do código que contextualizam e facilitam o aprendizado em cada etapa do processo. Este livro é recomendado a profissionais e alunos de pós-graduação que tenham interesse em aprender a utilizar o R. Não é necessário nenhum conhecimento prévio de programação para aproveitar o material. Após a leitura, é esperado que o leitor tenha o conhecimento suficiente para desenvolver suas própria pesquisa de forma autônoma, seja essa a confecção de documentos acadêmicos ou então o trabalho de pesquisa em empresas públicas e privadas. - Detail page: https://www.msperlin.com/book/2024-adfer - Read online: https://www.msperlin.com/books/adfeR/ - Amazon: https://www.amazon.com.br/dp/B08WNC27ZY ### Visualização de Dados com o R (2022) Visualização de dados é a arte de transformar medidas em padrões visuais. O objetivo é incrivelmente simples, apresentar evidências de um evento ou causalidade do mundo real. E, como qualquer arte, é fácil de descrever e difícil de executar. Na prática, construir figuras baseadas em dados exige expertise da área de conhecimento, uma noção de design e conhecimento das ferramentas corretas de programação. O retorno para o autor, porém, é imenso. Em um mundo digitalizado, gráficos baseados em dados são compartilhados com o mundo inteiro em questões de segundos, ao clique de um botão. Todo o material do livro, incluindo gráficos e código, é aberto ao público e reproduzível. Para facilitar a importação dos dados, construí um módulo do R chamado vdr, o qual pode ser instalado de forma prática e fácil. Este pacote contém todos os dados e funções usados no livro. Com ele será possível importar dados para a sua sessão do R e rodar todos os exemplos no seu próprio computador. - Detail page: https://www.msperlin.com/book/2022-vdr - Read online: https://www.msperlin.com/books/vdr/ - Amazon: https://www.amazon.com.br/dp/B0BK2V4HTB ### Introdução ao R - Um guia prático para iniciantes (2024) Guia prático e acessível voltado para iniciantes aprenderem a linguagem R e o RStudio para análise de dados, mesmo sem conhecimento prévio de programação. Cobre fundamentos, manipulação de dados com dataframes e conceitos fundamentais de programação estruturados com exemplos práticos reais do mercado financeiro e econômico. - Detail page: https://www.msperlin.com/book/2024-intror - Read online: https://www.msperlin.com/books/introR/ - Amazon: https://www.amazon.com.br/dp/B0CYN25PD6 ### Analyzing Financial and Economic Data with R (Third Edition) (2022) A comprehensive, practical guide designed to teach students, researchers, and data analysts how to use R and RStudio for professional data analysis in finance and economics. Follows a structured approach from basic setup to data acquisition, manipulation, visualization with ggplot2, and econometric modeling. - Detail page: https://www.msperlin.com/book/2023-afedr-en - Read online: https://www.msperlin.com/books/afedr/ - Amazon: https://www.amazon.com/dp/B0BZ98RYFW ### Questões de R em Concursos Públicos: Soluções e comentários (2023) Voltado para candidatos que precisam dominar a linguagem R para provas de concursos que exigem análise de dados. Apresenta uma curadoria de questões reais exigindo conhecimento de sintaxe e manipulação de dados, oferecendo soluções detalhadas e comentários explicativos sobre o raciocínio por trás de cada questão. - Detail page: https://www.msperlin.com/book/2023-qrcp - Read online: https://www.msperlin.com/books/qrcp/ - Amazon: https://www.amazon.com.br/dp/B0C6GZ6V8T ## Datasets ### GetDFPData2: Annual Financial Reports from the DFP System Annual financial statements (DFP) of Brazilian public companies, as published by B3/CVM. - Detail page: https://www.msperlin.com/data/getdfpdata2 - DOI: https://doi.org/10.7910/DVN/7VVX4J - Coverage: Annual financial reports of Brazilian public companies ### GetFREData: Corporate Datasets from the FRE System Corporate events and reference data from the FRE system (B3/CVM). - Detail page: https://www.msperlin.com/data/getfredata - DOI: https://doi.org/10.7910/DVN/QIMUNZ - Coverage: Corporate events and registration data of Brazilian public companies ### GetTDData: Prices and Yields of Brazilian Bonds Prices and yields of Brazilian sovereign bonds (Tesouro Direto). - Detail page: https://www.msperlin.com/data/gettddata - DOI: https://doi.org/10.7910/DVN/SCSQUF - Coverage: Brazilian sovereign bonds (Tesouro Direto) ## Peer-Reviewed Publications ### 2026 — The use of AI in 10-K filings: An empirical analysis - Authors: Marcelo S. Perlin, Cristian R. Foguesatto, Aliki K. Galanos, Felipe Affonso - Journal: Finance Research Letters - DOI: https://doi.org/10.1016/j.frl.2026.110161 - Detail page: https://www.msperlin.com/publications/2026-frl-ai-10ks - Markdown: https://www.msperlin.com/publications/2026-frl-ai-10ks.md With the rise on the use of AI in the corporate setting and the associated downside risk of bad disclose and overall credibility, we ask how much and which type of organizations rely on these tools when preparing their annual reports. Using a custom, publicly available, prediction model finetuned on FinBERT, we estimate the probability of AI-generated content and analyze the sentiment of three sections of the 10-K reports of Russell 1000 constituent companies from 2018 to 2024. Results indicate a relatively stable use of AI over the period, with distinct patterns across report sections, being the Business Description the section with the highest probability of AI adoption. Additionally, older, higher-risk (beta), and higher asset turnover firms appear more likely to adopt AI-assisted writing, in contrast to firms with higher accrual ratios and greater leverage. Moreover, there is a positive association between sentiment scores and AI detection within the Risk Factors section. These findings offer valuable insights for researchers who need to further examine AI-assisted writing in annual reports. They also carry important implications for practitioners such as investors and policy-makers who should pay closer attention to this emerging organizational practice. ### 2026 — The academia downturn: the impact of COVID-19 in the Brazilian research productivity - Authors: Marcelo S. Perlin, Denis Boreinstein, Takeyoshi Imasato, Cristian Foguesatto - Journal: Research Evaluation - DOI: https://doi.org/10.1093/reseval/rvag019 - Detail page: https://www.msperlin.com/publications/2026-re-covid - Markdown: https://www.msperlin.com/publications/2026-re-covid.md This study investigates the impact of the COVID-19 pandemic on research productivity in Brazil using a scientometrics approach. We analyze data from 134,492 researchers registered on the Lattes platform, focusing on publications, supervisions, research projects, and conference participation between 2010 and 2022. As expected, our findings reveal a significant decrease in productivity in all measures following the onset of the pandemic in 2020. Despite this overall decrease, there was a surge in research on COVID-19-related topics, particularly in the health sciences. We also found that, while gender did not significantly influence productivity changes, researchers with PQ scholarships and those with more than 19 years of experience since completing their PhD were disproportionately affected. These results highlight the need for policies to support these vulnerable groups during global crises. ### 2025 — Using ChatGPT for Creating Multi-Language Finance Related Sentiment Dictionaries - Authors: Marcos Reichert, Marcelo S. Perlin - Journal: Computational Economics - DOI: https://doi.org/10.1007/s10614-025-11233-3 - Detail page: https://www.msperlin.com/publications/2025-ce-using-chatgpt-for-creatin - Markdown: https://www.msperlin.com/publications/2025-ce-using-chatgpt-for-creatin.md The finance literature is abundant with applications of text sentiment based on English dictionaries. However, a large proportion of financial documents are written in languages other than English. As such, there is a gap in literature for the development of sentiment dictionaries in other languages. Using a reproducible and low cost approach based on ChatGPT’s API, and with minimal interventions, we attempt to fill this gap with the proposal of a methodology for building finance-related sentiment word lists for any language. We provide an empirical study by building a Portuguese finance-related sentiment dictionary with the proposed methodology, and using it to analyze the last 50 COPOM (Monetary Politics Committee) public statements. Furthermore, the dictionary’s performance is benchmarked against a full-text analysis NLP model, revealing a more balanced sentiment classification profile with a significant improvement in identifying positive news compared to existing lists. Additionally, the resulting sentiment word list in Portuguese is publicly available to the academic community and may help future local studies. ### 2025 — Can AI beat a naive portfolio? An experiment with anonymized data - Authors: Marcelo S. Perlin, Cristian Foguesato, Fernanda Muller, Marcelo Righi - Journal: Finance Research Letters - DOI: https://doi.org/10.1016/j.frl.2025.107126 - Detail page: https://www.msperlin.com/publications/2025-frl-can-ai-beat-a-naive-portf - Markdown: https://www.msperlin.com/publications/2025-frl-can-ai-beat-a-naive-portf.md Using anonymized data from the United States (U.S.) market, we evaluate the performance of Google’s main LLM (Large Language Model) Gemini 1.5 Flash in making investment decisions. Unlike other studies, we query the LLM for different investment horizons (1 to 36 months) and types of financial information (financial data, price data, and a combination of both). Running a total of 30,000 simulations for 1,522 companies over 20 years of data, we find that Gemini does not consistently outperform a naive portfolio and the S&P 500 index in terms of returns and Sharpe ratios. Additionally, our findings indicate a decline in risk adjusted investment performance as the investment horizon extends. ### 2025 — Profiling Brazil’s research elite: Insights from a cluster analysis of a large database - Authors: Cristian Foguesato, Denis Boreinstein, Marcelo S. Perlin, Takeyoshi Imasato - Journal: Journal of Data and Information Science - DOI: https://doi.org/10.2478/jdis-2025-0033 - Detail page: https://www.msperlin.com/publications/2025-jfis-profiling-brazils-resear - Markdown: https://www.msperlin.com/publications/2025-jfis-profiling-brazils-resear.md Purpose: This study analyzes the profiles of elite Brazilian researchers, recognized through the prestigious CNPq productivity scholarships. By identifying distinct researcher clusters, the study sheds light on different academic strategies, levels of productivity, and academic contributions within the Brazilian higher education system. Design/methodology/approach: The research analyzes a comprehensive dataset of 14,003 researchers, employing principal component analysis (PCA) followed by cluster analysis to group researchers based on their academic attributes. The clusters reflect diverse aspects of research productivity, graduate supervisions, and publication patterns. Findings: The analysis reveals the existence of three distinct researcher profiles (the Advanced Supervisors, the Book Publishers/Organizers, and the Generalists). The study also highlights the characteristics of high-caliber scientists, representing the upper echelon of Brazilian researchers in terms of productivity and impact. Research limitations: Although the study provides a robust analysis of the Brazilian system, the results reflect specific characteristics of the Brazilian academic context. Furthermore, the analysis was restricted to normalized annual data, which may overlook temporal variations in researcher productivity. Pratical implications: The findings provide valuable insights for policy makers, funding agencies (such as CNPq), and university administrators who aim to develop tailored support programs for different researcher profiles. Originality/value: The cluster-based profiling offers a novel perspective on how different academic trajectories coexist within a national science system, offering lessons for other emerging economies. ### 2025 — Understanding Financial Resilience through Innovation and Top Management - Authors: Aliki Galanos, Marcelo S. Perlin - Journal: Revista Brasileira de Gestão de Negócios - DOI: https://doi.org/10.7819/rbgn.v27i03.4302 - Detail page: https://www.msperlin.com/publications/2025-rbgn-understanding-financial - Markdown: https://www.msperlin.com/publications/2025-rbgn-understanding-financial.md Purpose – We aim to verify if and how top management’s acquired academic characteristics help firms withstand crises by examining the effect of CEOs’ and board members’ education and scholar expertise on financial resilience through innovation. Theoretical framework – We draw upon upper echelons, agency and information asymmetry theories. Design/methodology/approach – We collect data for the Brazilian stock market during the COVID-19 pandemic to allow observation in times of crisis. Using a unique dataset of academic curricula, we estimate OLS and Cox regressions to reach our goal and examine the stability and flexibility of firms during the pandemic. We perform robustness checks through quantile regression to provide more evidence for our findings. Findings – We find that top managements’ acquired academic characteristics have an impact on financial resilience. This relationship is sometimes moderated by innovation. Overall, innovation and top managements’ acquired academic characteristics seem advantageous during a crisis with their interaction boosting these relationships. Research Practical & Social implications – This study provides evidence of the importance of top management to yet another organizational outcome, i.e. resilience. Our contributions are both theoretical and empirical as we advance in the discussions about financial resilience, corporate governance and innovation while we deliver insights to practitioners and policymakers. Originality/value – We are the first to uncover some of the determinants of financial resilience in Brazil whose stock market tends to be more sensitive to shocks than that of more developed countries. Our study provides guidance to both investors and managers. ### 2024 — Risk Forecasting Comparisons in Decentralized Finance: An Approach in Constant Product Market Makers - Authors: Lucas Mussoi, Fernanda Maria Müller, Marcelo S. Perlin - Journal: Computational Economics - DOI: https://doi.org/10.1007/s10614-024-10585-6 - Detail page: https://www.msperlin.com/publications/2024-ce-risk-forecasting-comparisons-in-decentralized-finance - Markdown: https://www.msperlin.com/publications/2024-ce-risk-forecasting-comparisons-in-decentralized-finance.md This paper investigates risk measurement within the Decentralized Finance (DeFi) ecosystem, focusing on liquidity pools that utilize Constant Product Market Makers (CPMMs). The study evaluates and compares the performance of various market risk forecasting models, specifically Value-at-Risk (VaR) and Expected Shortfall (ES), applied to decentralized liquidity pool data. ### 2024 — Pricing efficiency in cryptocurrencies: The case of centralized and decentralized markets - Authors: Lucas Mussoi, Marcelo S. Perlin, Fernanda Maria Muller - Journal: Journal of Economics and Business - DOI: https://doi.org/10.1016/j.jeconbus.2024.106224 - Detail page: https://www.msperlin.com/publications/2024-jeb-pricing-efficiency-in-cry - Markdown: https://www.msperlin.com/publications/2024-jeb-pricing-efficiency-in-cry.md This article presents a comparative analysis of the Ethereum (ETH) weak form of market efficiency priced in Bitcoin (BTC), Dai (DAI), and Tether (USDT). The investigation encompasses data from Uniswap-V2, a decentralized app utilizing a Constant Product Market Maker (CFMM) for cryptocurrency pricing, and Binance, a centralized exchange. The study employs different rolling windows to apply the asymmetric MF-DFA. The efficiency of exchange pairs is ranked using the market deficiency measure (MDM). Besides aligning with the literature, revealing an efficiency increase with larger rolling window sizes across centralized and decentralized exchanges and overall, upward and downward trends, our findings reveal that the CFMM employed by Uniswap-V2 leads to a more efficient market for the ETH-BTC pair compared to Binance, making it among the first studies to compare efficiency across these exchanges types. To delve deeper into this phenomenon and explore the dynamics between distinct pricing mechanisms, the Thermal Optimal Path is employed. The analysis highlights a lead-lag relationship between ETH prices in centralized and decentralized exchanges. The results suggest that market efficiency emerges first in the decentralized exchange, particularly when ETH is priced in BTC. The asymmetric MF-DFA was also employed on the pairs datasets before and after the Ethereum 2.0 hard fork. The findings of this analysis revealed significant results indicating that following the fork, Uniswap-V2 exhibited superior market efficiency compared to Binance for the majority of overall and downward trends, a phenomenon that was not observed prior to the merge. These findings contribute to the existing literature on cryptocurrency market efficiency by emphasizing the influence of network upgrades in trading platforms. Notably, this research reveals that, for the ETH-BTC pair, decentralized exchanges exhibit superior a level of weak form efficiency. ### 2024 — The determinants and impact of research grants: The case of Brazilian productivity scholarships - Authors: Marcelo S. Perlin, Denis Boreinstein, Takeyoshi Imasato, Marcos Reichert - Journal: Journal of Informetrics - DOI: https://doi.org/10.1016/j.joi.2024.101563 - Detail page: https://www.msperlin.com/publications/2024-joi-the-determinants-and-impa - Markdown: https://www.msperlin.com/publications/2024-joi-the-determinants-and-impa.md Research Productivity Grant (PQ) is a governmental research award maintained by CPNq, the Brazilian Council of Research, and designed as a funding program to support scientific studies in all fields of science. Using a compilation of data from the Lattes platform, we study the individual CVs of more than 133000 researchers between 2005 and 2022 to examine PQ's selection criteria and impact on research productivity over time. First, a machine learning model can accurately predict who receives the financial support. This suggests that some parts of the evaluation process can be automated based on Lattes. Moreover, the main factors that impact the likelihood of a researcher receiving an entry-level PQ are the number of supervisions and papers published. These factors are consistent across different fields of science. Additionally, we found a significant and positive impact from receiving the award in key academic research output. After receiving a CNPq productivity award, researchers tend to increase the number of citations of papers and publications. ### 2023 — What is the sustainable withdraw rate for Brazil? - Authors: Lucas Pereira, Marcelo S. Perlin - Journal: Revista Brasileira de Finanças - DOI: https://doi.org/10.12660/rbfin.v21n3.2023.89040 - Detail page: https://www.msperlin.com/publications/2023-rbfin-what-is-the-sustainable-w - Markdown: https://www.msperlin.com/publications/2023-rbfin-what-is-the-sustainable-w.md The consistent increase in longevity and the decrease in birth rates in the Brazilian population have exacerbated the financial solvency of social security funds, threatening the retirement of a significant portion of the population. This study proposes an adaptation of the Trinity (Cooley et al., 1998) model for the Brazilian market, based on the ALM (Asset Liability Management) methodology and stochastic programming. It suggests using a personal investment portfolio as a source of funds during retirement. The study innovates by employing econometric models and simulation to address the typical problem of low data availability for local financial assets. The results indicate that a 5% withdrawal rate is sustainable and relatively safe for annual withdrawals from a portfolio composed primarily of fixed-income assets. ### 2022 — The Academic Inbreeding Controversy: Analysis and Evidence from Brazil - Authors: Marcelo S. Perlin, Denis Boreinstein, Takeyoshi Imasato - Journal: Journal of Informetrics - DOI: https://doi.org/10.1016/j.joi.2022.101287 - Detail page: https://www.msperlin.com/publications/2022-joi-the-academic-inbreeding-c - Markdown: https://www.msperlin.com/publications/2022-joi-the-academic-inbreeding-c.md This paper presents an embracing quantitative inbreeding analyses in the Brazilian higher education system (HES). Several studies were conducted about academic inbreeding in several countries with contradictory results on its effect in research productivity, indicating how controversial this issue is. This is the first comprehensive research based on data from more than 79,000 researchers from all fields of knowledge. We find that inbreeding can be found in all fields of science in Brazil. Results from a robust statistical analysis indicate that inbreds are significantly more productive than non-inbreds in all research publications, except in books. Particularly, we find that researchers that have spent a mobility period, either during doctoral studies or a scholarly visiting position, are more productive than other types of inbreds. The overall conclusion is that there is no evidence to support the detrimental view of academic inbreeding based on scientific productivity. We then discuss possible explanations to our findings and present suggestions of future research. ### 2022 — What drives the release of material facts for Brazilian stocks? - Authors: Marcelo S. Perlin, Marcos Henrique Reichert - Journal: Revista Brasileira de Finanças - DOI: https://doi.org/10.12660/rbfin.v20n2.2022.85378 - Detail page: https://www.msperlin.com/publications/2022-rbfin-what-drives-the-release-o - Markdown: https://www.msperlin.com/publications/2022-rbfin-what-drives-the-release-o.md In this study we look at the determinants of material facts (Fatos Relevantes) in the Brazilian Market. Following local legislation, material facts should be released to the public right after its occurrence, and preferably, after trading time. We investigate the randomness of the release of material facts—and release strategies by executives—and test whether there is a particular time period where more or fewer material facts are published. We also investigate whether the content of the material fact—positive or negative sentiment—explains different strategies regarding the release of news to the market. Lastly, using Vector Auto Regressions, we test for a feedback effect between material facts and the financial data, that is, if material facts publishing drives the returns, volume and volatility and vice versa. Finally, our results show that volume, volatility and returns (to a lesser extent) are determinants for material facts publishing and material facts. ### 2021 — O Impacto da Titulação Acadêmica de Conselheiros e Diretores sobre a Performance de Empresas Negociadas na B3 - Authors: Marcelo Perlin, Guilherme Kirch, Daniel Vancin, Mauro Mastella - Journal: Brazilian Business Review - DOI: https://doi.org/10.15728/bbr.2021.18.5.5 - Detail page: https://www.msperlin.com/publications/2021-bbr-o-impacto-da-titulacao-ac - Markdown: https://www.msperlin.com/publications/2021-bbr-o-impacto-da-titulacao-ac.md Estudos anteriores indicam que a titulação acadêmica dos conselheiros afeta a performance e o risco das empresas. No entanto, essas evidências não levam em conta detalhes sobre a formação e produção científica dos conselheiros acadêmicos. Este artigo investiga se a titulação acadêmica dos conselheiros e diretores impacta a performance de empresas negociadas em bolsa. A amostra compreendeu 133 empresas não financeiras, de 2010 a 2018, consolidando informações extraídas tanto da B3 como da plataforma Lattes. Os resultados demonstram que acadêmicos são mais comuns em conselhos e diretorias de maior tamanho e em empresas de controle público. Em geral, os modelos estimados indicam baixa relação entre a presença de acadêmicos e performance corporativa. Porém, observamos que a presença de doutores com títulos obtidos no exterior afeta positiva e significativamente a rentabilidade e negativamente o risco das empresas estudadas. Outros atributos acadêmicos do conselho apresentaram impacto sobre o risco das empresas, mas não sobre a sua rentabilidade. ### 2021 — Board gender diversity: performance and risk of Brazilian firms - Authors: Mauro Mastella, Daniel Vancin, Marcelo S. Perlin, Guilherme Kirch - Journal: Gender in Management - DOI: https://doi.org/10.1108/GM-06-2019-0088 - Detail page: https://www.msperlin.com/publications/2021-gm-board-gender-diversity-p - Markdown: https://www.msperlin.com/publications/2021-gm-board-gender-diversity-p.md Purpose: This study aims to intend to check if female board representation affects performance and risk and to analyse the evolution of the demographic aspects of the presence of women on boards in Brazil. Design/methodology/approach: The authors used a sample of 150 Brazilian publicly traded companies from 2010–2018, with different measures of firm performance, firm risk and women’s presence on the board. The study approach is based on a set of ordinary least squares, quantile and panel data regressions. Findings: The presence of women on the board has a positive effect on all of our accounting and market performance measures. However, the result of the impact on risk is not conclusive. The study also found that the number of females on the board has a more significant effect at the lower levels of firm performance measured by return on equity, but at the higher levels when measured by Tobin’s Q. Regarding return on assets, the more significant effect happened on the extremes of the performance distribution. The study findings point that market investors place more value in female presence on the board than in director positions. Originality/value: By estimating the impact of women’s presence on the boards of directors in firm performance and risk, this study aimed to verify this impact in different aspects of the company. In addition, the authors did so in a sample with many years, making it possible to evaluate the historical evolution of the feminine presence in the boards of administration as well as in the groups of directors, assisting Brazilian legislators with new evidence about the possible impacts of Draft Law 7179/2017. ### 2020 — Does algorithmic trading harm liquidity? Evidence from Brazil - Authors: Henrique P. Ramos, Marcelo S. Perlin - Journal: North American Journal of Economics and Finance - DOI: https://doi.org/10.1016/j.najef.2020.101243 - Detail page: https://www.msperlin.com/publications/2020-najef-does-algorithmic-trading - Markdown: https://www.msperlin.com/publications/2020-najef-does-algorithmic-trading.md This paper provides the first evidence of algorithmic trading (AT) reducing liquidity in the Brazilian equities market. Our results are contrary to the majority of work which finds a positive relationship between AT and liquidity. Using the adoption of a new data center for the B3 exchange as an exogenous shock, we report evidence that AT increased realized spreads in both firm fixed-effects and vector autoregression estimates for 26 stocks between 2017 and 2018 using high-frequency data. We also provide evidence that AT increases commonality in liquidity, evidencing correlated transactions between automated traders. ### 2020 — A GARCH Tutorial with R - Authors: Marcelo S. Perlin, Mauro Mastella, Daniel Vancin, Henrique P. Ramos - Journal: Revista de Administração Contemporânea - DOI: https://doi.org/10.1590/1982-7849rac2021200088 - Detail page: https://www.msperlin.com/publications/2020-rac-a-garch-tutorial-with-r - Markdown: https://www.msperlin.com/publications/2020-rac-a-garch-tutorial-with-r.md In this tutorial paper we will address the topic of volatility modeling in R. We will discuss the underlying logic of GARCH models, their representation and estimation process, along with a descriptive example of a real-world application of volatility modelling: we use a GARCH model to investigate how much time it will take, after the latest crisis, for the Ibovespa index to reach its historical peak once again. The empirical data covers the period between years 2000 and 2020, including the 2009 financial crisis and the current 2020’s episode of the COVID-19 pandemia. We find that, according to our GARCH model, Ibovespa is likely to reach its peak once again in 2,5 years. All data and R code used to produce this tutorial are freely available on the internet and all its results can be easily replicated. ### 2019 — Accessing Financial Reports and Corporate Events with GetDFPData - Authors: Marcelo S. Perlin, Guilherme Kirch, Daniel Vancin - Journal: Revista Brasileira de Finanças - Detail page: https://www.msperlin.com/publications/2019-rbfin-accessing-financial-repor - Markdown: https://www.msperlin.com/publications/2019-rbfin-accessing-financial-repor.md This paper presents and discusses the contributions and usage of GetDFPData, which is an open and free software for accessing corporate data from the Brazilian financial exchange, B3. The distribution and popularization of an open source algorithm for gathering and managing financial data can improve finance research and practice in two ways. First, by boosting the number and quality of research in accounting and corporate finance. Secondly, by providing retail investors with reliable data that may help their allocation decision. Initially, we analyzed the use of this kind of data in a list of recent publications to show the relevance of financial reports and corporate events data for research in the fields of accounting and finance. Finally, we illustrate the use of GetDFPData in large-scale research, an empirical and reproducible example of a corporate finance study. ### 2019 — A consumer credit risk structural model based on affordability: balance at risk - Authors: Marcelo S. Perlin, Marcelo Brutti Righi, Tiago P. Filomena - Journal: Journal of Credit Risk - Detail page: https://www.msperlin.com/publications/2019-a-consumer-credit-risk-st - Markdown: https://www.msperlin.com/publications/2019-a-consumer-credit-risk-st.md This paper introduces an approach designed for personal credit risk, with possible applications in risk assessment and optimization of debt contracts. We define a structural model related to the financial balance of an individual, allowing for cashflow seasonality and deterministic trends in the process. Based on the proposed model, we develop risk measures associated with the probability of default rates conditional on time. This formulation is best suited to short-term loans, where the dynamics of individuals’ cashflow, such as seasonality and uncertainty, can significantly impact future default rates. In the empirical section of this paper, we illustrate an application by estimating risk measures using simulated data. We also present the specific case of optimization of a financial contract, where, based on an estimated model, we find the yield rate/time to maturity pair that maximizes the expected profit or minimizes the default risk of a short-term debt contract. ### 2018 — Lotka’s law for the Brazilian scientific output published in journals - Authors: Sergio da Silva, Marcelo S. Perlin, Raul Matsushita, Takeyoshi Imasato, Denis Boreinstein - Journal: Journal of Information Science - DOI: https://doi.org/10.1177/0165551518801813 - Detail page: https://www.msperlin.com/publications/2018-jis-lotkas-law-for-the-brazi - Markdown: https://www.msperlin.com/publications/2018-jis-lotkas-law-for-the-brazi.md Lotka’s law is a power law for the frequency of scholarly publications. We show that Lotka’s law cannot be dismissed after considering a massive sample of the number of publications of Brazilian researchers in journals listed on the SCImago Journal Rank and the Journal Citation Reports. For the SCImago Journal Rank, we found a power law with the Pareto exponent of 0.4 beyond the threshold of 50 papers. This means computing the ‘average number of publications’ of either a researcher or a discipline is of no practical significance. ### 2018 — Is predatory publishing a real threat? Evidence from a large database study - Authors: Marcelo S. Perlin, Takeyoshi Imasato, Denis Boreinstein - Journal: Scientometrics - DOI: https://doi.org/10.1007/s11192-018-2750-6 - Detail page: https://www.msperlin.com/publications/2018-scientometrics-is-predatory-publishing-a - Markdown: https://www.msperlin.com/publications/2018-scientometrics-is-predatory-publishing-a.md Using a database of potential, possible, or probable predatory scholarly open-access journals, the objective of this research is to study the penetration of predatory publications in the Brazilian academic system and the profile of authors in a cross-section empirical study. Based on a massive amount of publications from Brazilian researchers of all disciplines during the 2000 to 2015 period, we were able to analyze the extent of predatory publications using an econometric modeling. Descriptive statistics indicate that predatory publications represent a small overall proportion, but grew exponentially in the last 5 years. Departing from prior studies, our analysis shows that experienced researchers with a high number of non-indexed publications and PhD obtained locally are more likely to publish in predatory journals. Further analysis shows that once a journal regarded as predatory is listed in the local ranking system, the Qualis, it starts to receive more publications than non-predatory ones. ### 2017 — Portfolio management using realized covariances: Evidence from Brazil - Authors: João F. Caldeira, Guilherme V. Moura, Marcelo S. Perlin, André A. P. Santos - Journal: EconomiA - DOI: https://doi.org/10.1016/j.econ.2017.05.001 - Detail page: https://www.msperlin.com/publications/2017-economia-portfolio-management-usin - Markdown: https://www.msperlin.com/publications/2017-economia-portfolio-management-usin.md It is often argued that intraday returns can be used to construct covariance estimates that are more accurate than those based on daily returns. However, it is still unclear whether high frequency data provide more precise covariance estimates in markets more contaminated from microstructure noise such as higher bid-ask spreads and lower liquidity. We address this question by investigating the benefits of using high frequency data in the Brazilian equities market to construct optimal minimum variance portfolios. We implement alternative realized covariance estimators based on intraday returns sampled at alternative frequencies and obtain their dynamic versions using a multivariate GARCH framework. Our evidence based on a high-dimensional data set suggests that realized covariance estimators performed significantly better from an economic point of view in comparison to standard estimators based on low-frequency (close-to-close) data as they delivered less risky portfolios. ### 2017 — Can we predict the financial markets based on Google's search queries? - Authors: Marcelo S. Perlin, João F. Caldeira, André A. P. Santos, Martin Pontuschka - Journal: Journal of Forecasting - DOI: https://doi.org/10.1002/for.2446 - Detail page: https://www.msperlin.com/publications/2017-jf-can-we-predict-the-financ - Markdown: https://www.msperlin.com/publications/2017-jf-can-we-predict-the-financ.md We look into the interaction of Google's search queries and several aspects of international equity markets. Using a novel methodology for selecting words and a vector autoregressive modeling approach, we study whether the search queries of finance-related words can have an impact on returns, volatility of returns and traded volume in four different English-speaking countries. We identify several words whose search frequency is associated with changes in the dependent variables. In particular, we find that increases in search queries including the word stock predict increased volatility and decreased index returns over the next week. On top of that, we investigate the performance of a market-timing strategy based on the search frequency of this word and benchmark it against random words from the Word-Net database and a naive buy-and-hold strategy. The results of this empirical application are positive and particularly stronger during the global crisis of 2009. ### 2017 — The Brazilian scientific output published in journals: A study based on a large CV database - Authors: Marcelo S. Perlin, André A. P. Santos, Takeyoshi Imasato, Denis Borenstein, Sergio da Silva - Journal: Journal of Informetrics - DOI: https://doi.org/10.1016/j.joi.2016.11.002 - Detail page: https://www.msperlin.com/publications/2017-ji-the-brazilian-scientific - Markdown: https://www.msperlin.com/publications/2017-ji-the-brazilian-scientific.md We assemble a massive sample of 180,000 CVs of Brazilian academic researchers of all disciplines from the Lattes platform. From the CVs we gather information on key variables that possibly explain the quantity and impact of their output published in journals: gender, PhD origin (domestic or foreign), time taken for finishing a PhD program, and number of years involved in research since PhD completion. Researcher productivity is gauged by three metrics: average SCImago Journal Rank impact factor, average points from Qualis (which is a Brazilian domestic metrics that includes domestic journals), and average number of journal publications per year after PhD completion. Taken together, we find males are more productive in terms of quantity of publications, but the effect of gender in terms of quality is mixed for individual groups of subject areas. Holding a PhD from abroad increases the chance for a researcher to publish in journals of higher impact, whereas domestic PhDs publish more articles, but in journals of less impact. Thus, there is a trade-off between research impact and quantity. We also find that the more years a researcher takes to finish his or her doctorate, the more likely he or she will publish less thereafter. The data also support the existence of an inverted U-shaped function relating research age and productivity. ### 2017 — Mispricing in the odd lots market in Brazil - Authors: Henrique P. Ramos, Marcelo S. Perlin, Marcelo B. Righi - Journal: North American Journal of Economics and Finance - DOI: https://doi.org/10.1016/j.najef.2017.08.006 - Detail page: https://www.msperlin.com/publications/2017-najef-mispricing-in-the-odd-lot - Markdown: https://www.msperlin.com/publications/2017-najef-mispricing-in-the-odd-lot.md We study the case of mispricing in the odd lots equity market in Brazil. Contrary to expectation, odd lot investors are paying higher prices than round lot investors. The pricing difference between markets is affected by market returns, volatility and spreads. Our main hypothesis is that; once the assets traded in the odd lot market are more illiquid than their counterparts, the mispricing is driven by liquidity factors. Additionally, we show that the mispricing yields an arbitrage opportunity that is not being traded away in the Brazilian market. Therefore, we propose regulators to review the market design for odd lots in Brazil. We argue that reducing the minimal trading unit in the round lots market would benefit investors. ### 2017 — Análise do Perfil dos Acadêmicos e de suas Publicações Científicas em Administração - Authors: Takeyoshi Imasato, Marcelo S. Perlin, Denis Borenstein - Journal: RAC. Revista de Administração Contemporânea (Impresso) - DOI: https://doi.org/10.1590/1982-7849rac2017150329 - Detail page: https://www.msperlin.com/publications/2017-rac-analise-do-perfil-dos-aca - Markdown: https://www.msperlin.com/publications/2017-rac-analise-do-perfil-dos-aca.md A produção da ciência tem sido foco de diversas discussões, particularmente ao que se refere à produtividade científica. Este artigo tem como objetivo analisar o perfil de acadêmicos doutores que atuam na área de Administração no Brasil e as suas respectivas publicações científicas, de modo a identificar fatores que expliquem esse aspecto da produtividade científica. Utilizando estatística descritiva e modelos estatísticos, com base em dados provenientes de 1976 pesquisadores doutores registrados na Plataforma Lattes, a pesquisa identificou fatores ligados à formação e experiência que influem na produtividade científica dos acadêmicos em Administração. Os resultados da pesquisa também indicam que houve importantes modificações nos padrões de formação acadêmica e de produtividade no país. Considerando a centralidade do tema da produtividade científica, o trabalho sugere novas pesquisas para que se ampliem as discussões sobre os rumos da produção intelectual e da pós-graduação em Administração no Brasil. ### 2017 — O poder preditivo de pesquisas na internet sobre o mercado financeiro brasileiro - Authors: Henrique Pinto Ramos, Kadja Katherine Mendes Ribeiro, Marcelo S. Perlin - Journal: RAM. Revista de Administração Mackenzie (Impresso) - DOI: https://doi.org/10.1590/1678-69712017/administracao.v18n2p184-210 - Detail page: https://www.msperlin.com/publications/2017-ram-o-poder-preditivo-de-pesq - Markdown: https://www.msperlin.com/publications/2017-ram-o-poder-preditivo-de-pesq.md Entender a capacidade preditiva de pesquisas no Google sobre o mercado financeiro brasileiro. Apesar de uma crescente literatura estrangeira utilizando dados sobre pesquisas oriundas no Google, no Brasil não se tem conhecimento de trabalhos desta natureza. A aplicação no mercado financeiro evidencia novas fontes de informação acerca do movimento dos mercados e pode contribuir para pesquisadores e praticantes entenderem melhor esta dinâmica. Principais aspectos metodológicos: Seguindo a literatura, foram estimados modelos VAR para investigar os efeitos em três variáveis dos mercados de renda acionário e de renda fixa: volume, retorno e volatilidade. Além disso, foi estudado o impacto da procura por termos relacionados à gripe no mercado de renda variável. Foram usados dados semanais de pesquisas do Google Trends e dos mercados financeiros entre o período de 2007 a 2014. Síntese dos principais resultados: Evidencia-se a existência de um efeito preditivo entre os níveis de pesquisas e as variáveis financeiras, principalmente no mercado de renda variável. Todavia, este resultado não foi robusto em todos os casos analisados. Destaca-se que, para a relação inversa, isto é, o mercado financeiro impactando as pesquisas no Google, encontrou-se forte evidência de uma relação causal. O uso de uma estratégia de negociação baseada neste tipo de dados gerou retornos maiores do que os bechmarkings definidos. O estudo revelou uma relação significativa entre o nível de pesquisas no Google e o mercado financeiro. Os resultados oferecem uma nova fonte de informação que afeta o mercado financeiro do Brasil. ### 2017 — Análise de Integração Financeira entre o Mercado Acionário Brasileiro e o Argentino: Uma Abordagem Dinâmica - Authors: Martin Pontuschka, Marcelo S. Perlin - Journal: Revista Brasileira de Finanças - Detail page: https://www.msperlin.com/publications/2017-rbfin-analise-de-integracao-fin - Markdown: https://www.msperlin.com/publications/2017-rbfin-analise-de-integracao-fin.md Neste artigo buscamos verificar a dinâmica da integração financeira entre as séries temporais do índice acionário brasileiro e argentino. Para isto, estimamos um modelo de estado de espaço através do filtro de Kalman, que permite a observação da dinâmica da integração financeira ao longo do tempo. Seguimos a proposta de Haldane e Hall (1991), que sugerem um modelo de espaço de estado para observar convergência entre diferentes séries temporais. Ao estimar o modelo conseguimos observar momentos de convergência e momentos de divergência entre o mercado acionário brasileiro e o argentino ao longo dos anos entre 1987 e 2014. Uma importante evidência foi a observação de divergência entre os índices acionários regionais em momentos de crise nos mercados internacionais. Ou seja, nos momentos de maior estresse nos mercados financeiros, o índice acionário brasileiro apresentou maior convergência em relação aos mercados internacionais do que em relação ao mercado argentino. Esta evidência está de acordo com Manning (2002), o qual verificou o mesmo efeito nos mercados acionários asiáticos durante a crise de 1997. ### 2016 — A Multistage Stochastic Programming Asset-Liability Management Model - An Application to the Brazilian Pension Fund Industry - Authors: Alan Delgado de Oliveira, Tiago Pascoal Filomena, Marcelo Scherer Perlin, Miguel Lejeune, Guilherme Ribeiro de Macedo - Journal: Optimization and Engineering - DOI: https://doi.org/10.1007/s11081-016-9316-3?wt_mc=internal.event.1.SEM.ArticleAuthorOnlineFirst - Detail page: https://www.msperlin.com/publications/2016-oe-a-multistage-stochastic-p - Markdown: https://www.msperlin.com/publications/2016-oe-a-multistage-stochastic-p.md This paper proposes a multistage stochastic programming approach for the asset-liability management of Brazilian pension funds. We generate asset price scenarios with stochastic differential equations—Geometric Brownian Motion model for stocks and Cox–Ingersoll–Ross model for fixed income securities. Intertemporal solvency regulatory rules for Brazilian pension funds are considered endogenously in the model and enforced with a combinatorial constraint. A VaR probabilistic constraint is incorporated to obtain a positive funding ratio at each time period with high probability. Our approach uses multiple trees to provide a representative characterization of the uncertainty and is not computationally prohibitive. We evaluate the insolvency probability under different initial funding ratios through extensive simulations. The study reveals that the likely decrease of interest rate premiums in the next years will force pension fund managers to significantly change their portfolio strategies. They will have to take more risk in order to deliver the cash flows required to cover the liabilities and satisfy the regulatory constraints. ### 2016 — A Microestrutura do Tesouro Direto: Sazonalidade do Fluxo de Ordens e a Formação de Spreads - Authors: Marcelo S. Perlin - Journal: Revista Brasileira de Finanças - DOI: https://doi.org/10.11606/1413-8050/ea140191 - Detail page: https://www.msperlin.com/publications/2016-rbfin-a-microestrutura-do-tesou - Markdown: https://www.msperlin.com/publications/2016-rbfin-a-microestrutura-do-tesou.md O objetivo deste artigo é analisar a microestrutura do Tesouro Direto em dois pontos: a dinâmica do fluxo de ordens e a formação dos spreads. Primeiro observa-se um padrão temporal da demanda por esses títulos na forma de uma sazonalidade em duas dimensões, dia do mês e mês do ano. No artigo descreve-se um modelo estatístico para a previsão da demanda com base nesta sazonalidade e analisa-se sua capacidade de previsão fora da amostra. Os resultados mostram que o modelo consegue capturar as propriedades das séries temporais, cuja previsibilidade pode auxiliar na administração do estoque de títulos por parte do governo. O segundo ponto de análise no estudo foi na formação do spread (preço de compra menos preço de venda) dentro do tesouro direto. No trabalho mostra-se que o Tesouro define o spread dos títulos negociados na plataforma do tesouro direto de acordo com regras simples sobre a maturidade de cada um dos instrumentos, formando um padrão do tipo escada. No desenvolvimento do artigo se discute os fatores negativos dentro desse método e apresenta-se também um método alternativo através de um modelo de ajustamento, o qual define o spread de forma mais parcimoniosa. ### 2016 — GetHFData: A R package for downloading and aggregating high frequency trading data from Bovespa - Authors: Marcelo S. Perlin, Henrique P. Ramos - Journal: Revista Brasileira de Finanças - Detail page: https://www.msperlin.com/publications/2016-rbfin-gethfdata-a-r-package-fo - Markdown: https://www.msperlin.com/publications/2016-rbfin-gethfdata-a-r-package-fo.md This paper introduces GetHFData, a R package for downloading, importing and aggregating high frequency trading data from the Brazilian financial market. Based on a set of user choices, the package GetHFData will download the required files directly from Bovespa’s ftp site and aggregate the financial data. The main objective of the publication of this software is to facilitate the computational effort related to research based on this large financial dataset and also to increase the reproducibility of studies by setting a replicable standard for data acquisition and processing. In this paper we present the available functions of the software, a brief description of the Brazilian market and several reproducible examples of usage. ### 2015 — A Estratégia de pares no Mercado Acionário Brasileiro: O Impacto da Frequência de Dados - Authors: Martin Pontuschka, Marcelo S. Perlin - Journal: RAM. Revista de Administração Mackenzie (Impresso) - DOI: https://doi.org/10.1590/1678-69712015/administracao.v16n2p188-213 - Detail page: https://www.msperlin.com/publications/2015-ram-a-estrategia-de-pares-no - Markdown: https://www.msperlin.com/publications/2015-ram-a-estrategia-de-pares-no.md A estratégia de pares é um popular método de negociação de ativos financeiros. Um dos motivos para isto se deve ao fato de que o resultado deste tipo de operação procede somente da relação entre os preços de dois ativos e não da direção do mercado. Somente a possibilidade de capturar ineficiências na precificação dos ativos é o que permite a obtenção de lucros sistemáticos através de um método de negociação de ativos. Com base na abertura de uma posição comprada (long), e no mesmo instante, uma posição vendida (short), esta estratégia de arbitragem estatística busca obter lucros por conta da convergência dos preços dos ativos negociados. Desta forma, o objetivo deste trabalho é analisar o desempenho da estratégia de pares em diferentes frequências de dados no mercado acionário brasileiro. Esta pesquisa baseou-se em Perlin (2007) onde foi encontrado o resultado de que as ineficiências de mercado aparecem em maior quantidade nos dados de maiores frequências, neste caso diário, semanal e mensal. O presente trabalho estende o leque de frequências entrando no universo intradiário com frequências de amostragem em 1 minuto, 5 minutos, 15 minutos, 30 minutos, 60 minutos e diários. O período de tempo da base de dados utilizada nesta pesquisa se estende de 01 de janeiro de 2008 até 31 de dezembro de 2011. Para compor a base de dados foram utilizados os vinte ativos com maior número de contratos negociados no período. A metodologia empregada nesta pesquisa utiliza a técnica de períodos de treinamento e períodos de negociações. Nos períodos de treinamento, os pares de ativos são selecionados de acordo com os menores desvios quadráticos nos seus preços normalizados. Nos períodos de negociação a estratégia verifica as operações realizadas em cada par de ativo previamente selecionado. Os resultados da pesquisa confirmam a hipótese primária de que quanto maior a frequência de amostragem, maiores evidências de ineficiência de mercado. Para chegar a esta conclusão foram comparados os índices de Sharpe da estratégia de pares nas diferentes frequências de dados. ### 2015 — Os pesquisadores, as publicações e os periódicos da área de Finanças no Brasil: Uma análise com base em currículos da plataforma Lattes. - Authors: Marcelo S. Perlin, André Portela Santos - Journal: Revista Brasileira de Finanças - Detail page: https://www.msperlin.com/publications/2015-rbfin-os-pesquisadores-as-publ - Markdown: https://www.msperlin.com/publications/2015-rbfin-os-pesquisadores-as-publ.md Este artigo analisa a produção científica da área de Finanças no Brasil. Utilizando um software proprietário para obter informações diretamente da plataforma de currículos Lattes foi possível verificar o perfil e as tendências da pesquisa em Finanças no território nacional. Os principais resultados da pesquisa mostram que a maioria dos pesquisadores de Finanças são relativamente jovens em termos de tempo de carreira, com doutorado finalizado entre os anos de 2005 e 2014 e situados no sudeste do país. Observa-se também que a produção científica nacional em periódicos internacionais de impacto é pequena em comparação com o total de publicações encontradas. O número de publicações por ano tem crescido exponencialmente, porém a qualidade das produções, medida pelo Qualis, deteriorou-se. Uma análise da produtividade dos autores mostra que os autores mais produtivos possuem duas características em comum: doutorado fora do Brasil e bolsa de produtividade do CNPQ. ### 2014 — On the Performance of the Tick Test - Authors: Marcelo S. Perlin, Chris Brooks, Alfonso Dufour - Journal: Quarterly Review of Economics and Finance - DOI: https://doi.org/10.1016/j.qref.2013.07.001 - Detail page: https://www.msperlin.com/publications/2014-qref-on-the-performance-of-the - Markdown: https://www.msperlin.com/publications/2014-qref-on-the-performance-of-the.md In financial research, the sign of a trade (or identity of trade aggressor) is not always available in the transaction dataset and it can be estimated using a simple set of rules called the tick test. In this paper we investigate the accuracy of the tick test from an analytical perspective by providing a closed formula for the performance of the prediction algorithm. By analyzing the derived equation, we provide formal arguments for the use of the tick test by proving that it is bounded to perform better than chance (50/50) and that the set of rules from the tick test provides an unbiased estimator of the trade signs. On the empirical side of the research, we compare the values from the analytical formula against the empirical performance of the tick test for fifteen heavily traded stocks in the Brazilian equity market. The results show that the formula is quite realistic in assessing the accuracy of the prediction algorithm in a real data situation. ### 2013 — The determinants of a cross market arbitrage opportunity: theory and evidence for the European bond market - Authors: Marcelo S. Perlin, Alfonso Dufour, Chris Brooks - Journal: Annals of Finance - DOI: https://doi.org/10.1007/s10436-013-0242-5 - Detail page: https://www.msperlin.com/publications/2013-af-the-determinants-of-a-cross-market-arbitrage - Markdown: https://www.msperlin.com/publications/2013-af-the-determinants-of-a-cross-market-arbitrage.md This paper investigates the factors that drive the occurrence of arbitrage opportunities across different trading platforms in fixed income markets. We develop a structural model that decomposes the likelihood of a cross-market arbitrage opportunity into fixed costs of trading, price update delays, and order flow impact, and test it empirically using European bond market data. ### 2013 — Comunalidades na Liquidez – Evidências e Comportamento Intradiário para o Mercado Brasileiro - Authors: Fernanda Gomes Victor, Marcelo S. Perlin, Mauro Mastella - Journal: Revista Brasileira de Finanças - Detail page: https://www.msperlin.com/publications/2013-rbfin-comunalidades-na-liquidez - Markdown: https://www.msperlin.com/publications/2013-rbfin-comunalidades-na-liquidez.md O objetivo deste artigo é estudar a dinâmica da liquidez intradiária na Bolsa de Valores Brasileira, sob a ótica de co-movimentos (ou comunalidades). No trabalho argumenta-se que este fator comum na liquidez das ações é afetado pelos efeitos intradiários particulares da microestrutura do mercado. Utilizando uma base de dados de alta frequência e o volume negociado como proxy para liquidez, tal hipótese é investigada para os dados brasileiros, sendo encontrada forte evidência de que o efeito de comunalidade da liquidez muda de forma significativa ao longo de diferentes intervalos do dia. Durante as primeiras e últimas horas de negociação este efeito é mais intenso, justificando-se tal resultado como um produto da chegada de novas informações ao mercado e uma consequência do risco overnight. ### 2013 — Os Efeitos da Introdução de Agentes de Liquidez no Mercado Acionário Brasileiro - Authors: Marcelo S. Perlin - Journal: Revista Brasileira de Finanças - Detail page: https://www.msperlin.com/publications/2013-rbfin-os-efeitos-da-introducao - Markdown: https://www.msperlin.com/publications/2013-rbfin-os-efeitos-da-introducao.md O objetivo principal deste artigo é analisar empiricamente o efeito da introdução de formadores de mercado no processo de compra e venda de ações no mercado acionário brasileiro. Agregando informações sobre as datas de início de contrato de formadores de mercado a uma privilegiada base de dados de alta frequência, foi possível efetivar um estudo de evento para verificar os efeitos da introdução dos agentes de liquidez. Conforme esperado, observou-se que o formador de mercado aumentou significativamente a liquidez dos papéis. Foi possível verificar um aumento médio de 31% no número de negócios entre o período anterior e posterior ao início das atividades do formador de mercado. Reporta-se também que a entrada do agente de liquidez muda significativamente a autocorrelação dos sinais de transação em aproximadamente 10%, sendo este resultado mais forte para as ações menos líquidas. Além disso, a pesquisa também indica resultados heterogêneos para a performance do formador de mercado quando os mesmos são segmentados pela instituição financeira de origem. Tal informação pode servir de base para uma escolha inteligente do formador de mercado por parte das empresas negociadas em bolsa. ### 2011 — System-Wide liquidity risk in the UK’s large-value payment system: an empirical analysis - Authors: Marcelo S. Perlin, Jochen Schanz - Journal: Bank of England Working Paper Series - Detail page: https://www.msperlin.com/publications/2011-bewps-system-wide-liquidity-ris - Markdown: https://www.msperlin.com/publications/2011-bewps-system-wide-liquidity-ris.md When settling their own liabilities and those of their clients, settlement banks rely on incoming payments to fund a part of their outgoing payments. We investigate their behaviour in CHAPS, the United Kingdom’s large-value payment system. Our estimates suggest that in normal times, banks increase their payment outflows when their liquidity is above target and immediately following the receipt of payments. We use these estimates to determine the robustness of this payment system to two hypothetical behavioural changes. In the first, a single bank stops sending payments, perhaps because of an operational problem. In the second, it pays out exactly what it previously received, relying exclusively on the liquidity provided by other system members. Using the observed uncertainty around our estimated behavioural equations, we derive probabilistic statements about the time at which the bank’s counterparties would run out of liquidity if they followed their estimated normal-time behaviour. ### 2009 — Evaluation of pairs-trading strategy at the Brazilian financial market - Authors: Marcelo S. Perlin - Journal: Journal of Derivatives and Hedge Funds - DOI: https://doi.org/10.1057/jdhf.2009.4 - Detail page: https://www.msperlin.com/publications/2009-jdhf-evaluation-of-pairs-tradi - Markdown: https://www.msperlin.com/publications/2009-jdhf-evaluation-of-pairs-tradi.md Pairs-trading is a popular trading strategy that tries to take advantage of market inefficiencies in order to obtain profit. The idea is simple: find two stocks that move together and take long/short positions when they diverge abnormally, hoping that the prices will converge in the future. From the academic point of view of weak market efficiency theory, pairs-trading strategy should not present positive performance, as, according to it, the actual price of a stock reflects its past trading data, including historical prices. This leaves us with a question: does pairs-trading strategy present positive performance for the Brazilian market? The main objective of this research is to verify the performance and risk of pairs-trading in the Brazilian financial market for different frequencies of the database: daily, weekly and monthly prices for the same time period. The main conclusion of this simulation is that pairs-trading strategy was a profitable and market-neutral strategy at the Brazilian market. Such profitability was consistent over a region of the strategy's parameters. The best results were found for the highest frequency (daily), which is an intuitive result. ### 2007 — Non-Linear Modelling in the Brazilian Market: Evaluating the Forecasting Performance of NN Univariate Nearest Neighbor and SNN Simultaneous Nearest Neighbour Forecasting Algorithm - Authors: Marcelo S. Perlin, Paulo S. Ceretta - Journal: REAd. Revista Eletrônica de Administração (Porto Alegre. Online) - Detail page: https://www.msperlin.com/publications/2007-read-non-linear-modelling-in-the-brazilian-market - Markdown: https://www.msperlin.com/publications/2007-read-non-linear-modelling-in-the-brazilian-market.md This paper presents an empirical evaluation of non-linear forecasting models applied to the Brazilian financial market. The study compares the predictive performance of Univariate Nearest Neighbor (NN) and Simultaneous Nearest Neighbour (SNN) forecasting algorithms in financial time series. ### 2006 — Teoria do Caos aplicada aos Contratos de Café no Mercado de Derivativos - Authors: Marcelo S. Perlin, Paulo S. Ceretta - Journal: READ - Revista Eletrônica da Administração (UFRGS) - Detail page: https://www.msperlin.com/publications/2006-read-teoria-do-caos-aplicada-aos-contratos-de-cafe - Markdown: https://www.msperlin.com/publications/2006-read-teoria-do-caos-aplicada-aos-contratos-de-cafe.md Este artigo investiga a presença de comportamento caótico e dinâmicas não lineares em contratos futuros de café negociados no mercado de derivativos brasileiro. Através da aplicação de testes quantitativos e qualitativos para sistemas não lineares, os resultados encontram evidências favoráveis à existência de comportamento caótico, com implicações diretas sobre a limitação da previsibilidade dos preços no longo prazo e a hipótese de eficiência de mercado.