News sentiment and international equity markets during BREXIT period: A textual and connectedness analysis

Abstract

This study used textual analysis of 34,209 news articles to quantify news sentiment into three main clusters-positive, negative and neutral-before analysing how they co-move with international equity indices, using the time-varying connectedness of Yilmaz (2009, 2012). Better understanding of the spillover of news sentiment to stock markets could aid the decision-making of institutional investors when strong uncertainty is present across major economies. We found that limited spillover from news sentiment to equity markets existed for both the European and international indices examined in the analysis, with spillover being stronger among smaller subsets of news articles more relevant for financial market participants. Additionally, the results indicated that, in the full sample, directional spillover was especially strong in times of larger uncertainty concerning BREXIT developments, whereas the smaller subsets, although also displaying stronger spillover during BREXIT uncertainty, revealed additional spillover peaks at times less related to major BREXIT developments. Differentiation between news about UKbased and EU-based companies also showed less spillover from news sentiment regarding EU-based companies, possibly implying that investors saw BREXIT developments as less relevant for the latter.

Does media sentiment affect stock prices? Evidence from China’s STAR market

“…In other words, the tone or sentiment of media coverage can affect investors’ judgment and, ultimately, asset pricing. Further, the relationship between media sentiment and prices will be strengthened in times of larger uncertainty concerning external information ( Koch et al, 2022 ).…”

Machine learning sentiment analysis, COVID-19 news and stock market reactions

“… Xu et al (2022) propose a news-based manager sentiment that is built on the tone of managers’ news reports and find strong predicting ability in returns both in- and out-of-sample. Koch et al (2022) study the impact of news sentiment in more than 34,000 news articles concerning Brexit and show the existence of limited spillover from news sentiment to equity markets. Nofer and Hinz (2015) show that weighted social mood levels on Twitter also have predictive value to stock market developments.…”