An Econometric Analysis of the S&P 500 with the Sentiments of President Trump's Tweets in his First Term

In today's digital era, social media transcends mere communication, significantly influencing public opinion and societal trends, notably in politics. Platforms like Twitter, now X, allows politicians to directly connect with a global audience, leading to profound implications. This shift is especially impactful in the financial sector, where social media-shared information can swiftly affect market trends and alter public perceptions, highlighting its pivotal role in modern societal dynamics. Our study examines the impact of political sentiments expressed in President Donald Trump's Tweets during his first term on financial markets. We use machine learning to analyze and calculate the sentiments of the President's Tweets from 2014 to 2020. We then use econometric models to assess the impact of the sentiments on the S&P 500 index fluctuations. A mixed-method approach was employed, combining sentiment analysis and topic modeling to study the tone and thematic content of the Tweets. To establish a causal relationship between the Tweets and the S&P 500 index movements, we use Vector Autoregression (VAR), Granger Causality Test, and Impulse Response Function (IRF). These methods allowed for a comprehensive assessment of both the temporal correlation and the causal links between the Tweets and market responses. Initial findings show a notable link between the sentiments and the content of the Tweets and short-term S&P 500 movements. The market reacted to both the factual content and the sentiment of the Tweets, with certain topics eliciting positive responses and other negative, significantly influencing market trends. The findings of this study underscore the significant impact that political figures can have on financial markets through social media platforms.