In this project, I utilize Twitter API to identify if a casual relationship exists between COVID-19 and movements in the stock market. The goal of the analysis was to show how Natural Language Processing and Text Analysis could be used to study the effects of COVID on stock market price action.

The analysis begins by providing an overview of NLP and then proposes a theory for why NLP can be used in financial markets.

Data is collected from Twitter that can be categorized as COVID related. Text analysis is performed on this data. Stock data was collected over a period preceding the spread of COVID (Nov. 2019) as well as a period after the first COVID case was announced (Dec. 2019) for each stock in the S&P 500. End of Day data was used. Metrics and a benchmark was selected to compare the pre and post COVID periods effect on market activity.

Methods were created to randomly select tweets and stock trading activity on the day of the tweet and make a comparison to that stock's benchmark trading activity.