Using sentiment analysis to evaluate the impact of the COVID-19 outbreak on Italy’s country reputation and stock market performance

During the recent Coronavirus disease 2019 (COVID-19) outbreak, the microblogging service Twitter has been widely used to share opinions and reactions to events. Italy was one of the frst European countries to be severely afected by the outbreak and to establish lockdown and stay-at-home orders, potentially leading to country reputation damage. We resort to sentiment analysis to investigate changes in opinions about Italy reported on Twitter before and after the COVID-19 outbreak. Using diferent lexicons-based methods, we fnd a breakpoint corresponding to the date of the frst established case of COVID-19 in Italy that causes a relevant change in sentiment scores used as a proxy of the country’s reputation. Next, we demonstrate that sentiment scores about Italy are associated with the values of the FTSE-MIB index, the Italian Stock Exchange main index, as they serve as early detection signals of changes in the values of FTSE-MIB. Lastly, we evaluate whether diferent machine learning classifers were able to determine the polarity of tweets posted before and after the outbreak with a diferent level of accuracy.