Vieira, Pedro RinoKorompli, Sofia2025-04-012025-04-012024-10Korompli, Sofia (2024). “Sentiment analysis of financial news on Bloomberg and its impact on the euro Stoxx 50”. Dissertação de Mestrado. Universidade de Lisboa. Instituto Superior de Economia e Gestãohttp://hdl.handle.net/10400.5/99877In today’s fast-paced financial markets, where investors are flooded with huge amounts of news and data, understanding the role of market sentiment has become increasingly important. This work investigates the impact of financial news sentiment on the daily returns of the Euro Stoxx 50 index, covering the period from January 2022 to March 2024. By examining whether sentiment extracted from Bloomberg articles can predict stock returns, the study aims to shed light on how investor psychology, driven by news, interacts with market performance, especially when accounting for market volatility (VIX) and lagged returns. To explore this relationship, sentiment scores were calculated using the Loughran-McDonald Lexicon and combined with stock price data to conduct panel regressions. Both fixed effects and random effects models were considered, with the Hausman test used to select the appropriate model. Additionally, OLS regressions for each year were applied to further understand the dynamics between sentiment and returns over time. The results reveal that in 2022, sentiment had a significant positive effect on returns, suggesting that investor sentiment played a key role during periods of heightened market uncertainty. The findings indicate that while sentiment can affect market performance, its impact is largely contingent on volatility levels and broader market conditions. This study contributes to the growing body of literature on the interaction between sentiment and financial markets, emphasizing the variable nature of sentiment’s influence across different market environmentsNos mercados financeiros dinâmicos de hoje, onde os investidores são inundados com grandes quantidades de notícias e dados, compreender o papel do sentimento de mercado tornou-se cada vez mais importante. Este trabalho investiga o impacto do sentimento das notícias financeiras nos retornos diários do índice Euro Stoxx 50, cobrindo o período de janeiro de 2022 a março de 2024. Ao examinar se o sentimento extraído de artigos da Bloomberg pode prever os retornos das ações, o estudo busca esclarecer como a psicologia dos investidores, impulsionada pelas notícias, interage com o desempenho do mercado, especialmente ao levar em conta a volatilidade do mercado (VIX) e os retornos defasados. Para explorar essa relação, as pontuações de sentimento foram calculadas usando o Loughran-McDonald Lexicon e combinadas com dados de preços de ações para realizar regressões em painel. Foram considerados modelos de efeitos fixos e efeitos aleatórios, com o teste de Hausman utilizado para selecionar o modelo mais adequado. Além disso, foram aplicadas regressões OLS para cada ano, a fim de entender melhor as dinâmicas entre sentimento e retornos ao longo do tempo. Os resultados revelam que, em 2022, o sentimento teve um efeito positivo significativo nos retornos, sugerindo que o sentimento dos investidores desempenhou um papel crucial durante períodos de maior incerteza no mercado. Os achados indicam que, embora o sentimento possa afetar o desempenho do mercado, seu impacto é amplamente condicionado pelos níveis de volatilidade e pelas condições gerais do mercado. Este estudo contribui para o crescente corpo de literatura sobre a interação entre sentimento e mercados financeiros, enfatizando a natureza variável da influência do sentimento em diferentes ambientes de mercado.engFinancial newsSentiment analysisEurostoxx 50Market efficiencyBehavioral financeStock returnsNotícias financeirasAnálise de sentimentoEurostoxx 50Eficiência de mercadoFinanças comportamentaisRetornos de açõesSentiment analysis of financial news on Bloomberg and its impact on the euro Stoxx 50master thesis