SOCIAL MEDIA INFLUENCE ON PETROBRAS’ VALUE

Abstract

This research seeks to contribute to the discussion on investor sentiment. The objective of the research was to analyze the relationship between social media sentiment on Twitter and the return of Petrobras. Thus, tweets about Petrobras, from 2010 to 2020, were analyzed to verify whether the textual sentiment of these messages impacts company return. The method to classify the words was the perceptual mapping together with the mean and standard deviation of the frequency of the terms in positive or negative days. The results showed that the variation of the total sentiment and the negative sentiment are related to the return. The variation in sentiment is significant when the market is bearish, demonstrating that the negative variation in sentiment intensifies the bearish movement in the market, smoothing the market’s decline when the variation in sentiment is positive. The findings corroborate the behavioural finance theory that sentiment is related to return, textual feeling in the case of the study. Finally, different samples were analyzed using the same methodology. With this, it was possible to verify that individuals talk more about the state-owned company than about the stock tickers or about the company itself on its official page.